Clinical value and cost-effectiveness analysis of conditionally approved anticancer drugs in China from 2015 to 2023.

Highlight box

Key findings

• Among 58 first-approved anticancer drugs in China (2015–2023), 20 (34.5%) received conditional approval. Conditionally approved drugs demonstrated comparable objective response rates to non-conditionally approved drugs, with no significant differences in overall survival or progression-free survival. Confirmatory trials of 23 converted drugs showed consistent efficacy outcomes. However, no correlation existed between clinical benefits and average daily treatment costs, revealing a pricing-value misalignment.

What is known and what is new?

• China’s conditional approval pathway expedites access to urgently needed therapies, but clinical value and evidentiary gaps remain uncertain.

• This study demonstrates that conditionally approved drugs achieve similar efficacy despite relying on surrogate endpoints, while revealing significant pricing-value disconnection requiring mechanism refinement.

What is the implication, and what should change now?

• Regulatory agencies should strengthen post-approval confirmatory study requirements while balancing clinical urgency with evidence robustness. Healthcare systems should reform pricing mechanisms to align innovative drug costs with demonstrated therapeutic benefits, ensuring sustainable patient access.

Introduction

The U.S. Food and Drug Administration (FDA) first established the accelerated approval (AA) pathway in 1992, primarily as a response to the urgent need for effective treatments during the Acquired Immunodeficiency Syndrome (AIDS) crisis ( 1 ), which was later employed for cancer drugs to expedite drug approvals for severe and life-threatening diseases ( 2 ). This process permitted pharmaceutical companies to employ surrogate endpoints as a measure to evaluate the anticipated therapeutic efficiency of treatments ( 3 ). Since 2015, China had also introduced a series of AA pathways, such as conditional approval (CA), breakthrough therapy, and priority review ( 4 - 7 ), to encourage clinical value-oriented drug innovation and accelerate the availability of clinically urgently needed drugs. In July 2020, with the implementation of the new version of the measures for the “Administration of Drug Registration”, the CA process was officially established. The 2015 reforms marked the policy foundation for expedited pathways, while the 2020 regulations provided formal legal codification and operational framework. It was primarily applied to drugs for severely life-threatening diseases with no effective treatments, as well as drugs urgently needed for public health emergencies, which applies to both novel drugs and supplementary cancer indications. Of note, this is similar to the U.S. AA pathway, which also covers both initial and supplementary indications, whereas the European conditional marketing authorization is restricted to initial approvals only. In this study, we focused exclusively on novel anticancer drugs and did not include supplementary cancer indications.

Under this system, the National Medical Products Administration (NMPA) could approve the marketing of drugs based on surrogate endpoints or early clinical trial data under certain conditions, thus shortening the drug development cycle ( 8 , 9 ). Similar to the U.S. AA pathway, marketing authorization holders are required to complete confirmatory trials within stipulated timeframes (typically specified in approval conditions) and submit periodic progress reports to the Center for Drug Evaluation (CDE). If confirmatory trials fail to demonstrate the expected clinical benefit or are not completed within the required period, the NMPA reserves the authority to revoke the marketing authorization.

During the approval process, NMPA would consider the risk-benefit ratio of drugs comprehensively, and drugs could be approved when the benefits outweigh the risks ( 10 ). After the drug approved by the NMPA, whether to include the drug in the National Reimbursement Drug List (NRDL) or not will be decided by the National Healthcare Security Administration (NHSA) based on its clinical value and cost-effectiveness compared to existing standard therapies, and the medical insurance payment standard will be determined through national healthcare insurance negotiation ( 11 ).

However, a series of controversies have been raised over the AA pathway established by FDA recently. Madhusoodanan et al. ( 3 ) reported in August 2023 that currently approximately 200 cancer treatments had been approved via AA, while 13.0% (n=26) of cancer drugs were withdrawn due to failed post-approval trials (median withdrawal time: 4 years; sometimes it takes decades) and 34.0% (n=68) were still awaiting confirmatory evidence. Beyond withdrawal rates, the AA pathway faces multiple systemic challenges. First, evidence quality concerns persist, as many confirmatory trials continue to rely on surrogate endpoints rather than definitive clinical outcomes, creating an ongoing evidence gap particularly for me-too drugs ( 12 , 13 ). Second, significant procedural delays have been documented both in completing post-approval confirmatory trials ( 14 ) and in withdrawing indications when clinical benefit is not confirmed ( 15 ), with approximately one in five indications withdrawn over the past decade ( 16 ). Third, safety concerns have emerged, as drugs approved via AA demonstrate more frequent postmarketing safety actions compared to those approved through standard pathways ( 17 ). Finally, the current system shows limited ability to discriminate clinical value ( 14 ), with rare likelihood that new evidence of meaningful clinical benefit will emerge after conversion from accelerated to full approval ( 12 ).

Currently, few studies evaluate the specific clinical benefits of CA drugs in China and their cost-effectiveness relationship, thereby the relevant situation of CA drugs is unclear ( 18 ). In 2023, Luo et al. evaluated the pre-approval clinical trial evidence of CA drugs approved from 2015 to 2022 and indicated that CA had markedly improved access to oncology therapies in clinical practice, though its reliance on inadequately validated surrogate endpoints poses persisting uncertainties regarding clinical benefit verification ( 19 ).

Nevertheless, owing to limited data availability, that study did not incorporate more comprehensive clinical benefit assessments using overall survival (OS) and progression-free survival (PFS), which are recognized as the most critical indicators for evaluating clinical benefits. Moreover, the cost-effectiveness of CA drugs and whether gaps existed between pre-approval and confirmatory clinical trials remained unexplored. To address these gaps, this retrospective study had two main objectives. The primary objective was to systematically evaluate the clinical value of anticancer drugs approved via both CA and non-CA pathways in China from 2015 to 2023, comparing clinical benefits through multidimensional indicators including objective response rate (ORR), OS, and PFS based on pivotal clinical trial evidence for each drug. Given that the CA pathway permits market entry based on early-phase, surrogate-endpoint-driven evidence, a key question is whether this expedited access compromises the clinical value of approved drugs. We hypothesized that anticancer drugs approved via the CA pathway do not demonstrate inferior clinical benefit compared with non-CA drugs. To test this hypothesis, we used non-CA drugs approved during the same period as a benchmark, comparing trial design characteristics, clinical benefit scores, and treatment costs between the two groups. In addition, we tracked the post-marketing confirmatory trial outcomes of CA drugs to assess whether clinical benefit was confirmed after approval.

The secondary objective was to explore the relationship between clinical benefit and the average daily cost of treatment. In this study, clinical benefit was defined as measurable improvements in efficacy endpoints derived from pivotal clinical trials, including ORR, OS, and PFS. Clinical value was defined as a broader concept integrating clinical benefit with economic considerations, specifically the relationship between efficacy outcomes and the average daily cost of treatment. We present this article in accordance with the CHEERS reporting checklist (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2026-1-0038/rc ).

Methods

Data sources and extraction

In this retrospective study, we extracted a list of approved anticancer drugs for the first time in China, and determined the basic information such as the approval date, approval indications, and approval pathways for the drugs on the list by the official websites of the NMPA ( 20 ) and its CDE ( 21 ), as well as the Pharmacodia database ( 22 ), covering the period from 2015 to 2023. We also obtained the sales prices at the time of approval and the latest sales prices of drugs from the Yilian Pharmaceutical Procurement Database ( 23 ) and provincial centralized pharmaceutical procurement platforms. Since no publicly available database exclusively lists novel anticancer drugs in China, our screening process began with the full NMPA drug approval database. The term “new drugs” in this study refers to new chemical entities and new biological products, excluding chemical generics, vaccines, blood products, biosimilars, new indications, new dosage forms, new routes of administration and changes in the applicable population, and other scenarios. These categories were excluded at the initial screening stage to identify novel drug entities, regardless of whether they pertained to anticancer products or were eligible for the CA pathway. In addition, this study excluded cellular therapy products, radiotherapy drug-device combinations, and drugs without review reports.

Ethics

Since we utilized publicly accessible data only, Institutional Review Board approval was unnecessary. Data for this study were obtained from literature and review reports, so patient informed consent was not required.

Measurement of clinical benefits

To evaluate the clinical outcomes of drugs with the same indications, we derived pivotal clinical trials information supporting drug approval from the technical review reports of drugs published by CDE, based on which we further searched for peer-reviewed trial publications to collect the latest results of these pivotal clinical trials supporting drug approval. We then conducted a pooled analysis of the main efficacy outcomes from these pivotal clinical studies. For randomized clinical trials (RCTs), clinical benefit of the investigated therapy was evaluated using pooled hazard ratio (HR) of OS and/or PFS. For non-randomized trials, clinical benefit of the investigated therapy was measured using pooled value of point estimates [ORR or complete remission/complete remission with partial hematological recovery (CR/CRh)] of the study primary endpoints. To compare clinical benefits between CA and non-CA drugs, and between the same drugs in pre-approval pivotal trials and post-approval confirmatory trials, differences between HR for OS and/or PFS or points estimates of ORR were also calculated. For trials reporting OS or PFS with a comparator arm, HR <0.75 with P<0.05 was used to indicate clinically and statistically significant improvements. For single-arm trials, clinical benefit was assessed based on whether the ORR threshold was met as reported in the published trial results or regulatory review documents.

Calculation of treatment cost

In this study, “cost-effectiveness analysis” refers to the relationship between average daily treatment cost and clinical benefit indicators (ORR, PFS, and OS). Based on the latest sales prices of drugs from the database mentioned above, this study further obtained the dosage information from the drug labels, and calculated the average daily treatment cost of each anticancer drugs. Thereby, the drug prices in this study refer to daily treatment cost. For dosages based on body weights or body surface areas, this study applied an average body weight of 70 kg and a body surface area of 1.7 m 2 ( 24 , 25 ). For a specific number of days in a treatment cycle, the average daily dosage for that cycle was calculated. When there is an initial dosage and maintenance dosage in a treatment cycle, the maintenance dosage was selected for calculation. For drugs with multiple approved dosages, the dosage with the lowest cost per unit was adopted ( 26 ).

Statistical analysis

Descriptive statistics were used to describe the characteristics of the new anticancer drugs first approved in China. For the primary efficacy endpoints from pivotal clinical studies (ORR for single-arm trials; HR for PFS and/or OS for RCTs), we performed a pooled analysis using meta-analysis to estimate the clinical benefits of CA and non-CA drugs. Statistical heterogeneity was evaluated using I 2 statistic {calculated by Cochran’s Q [100 × (Q – df)/Q]} and χ 2 ( 27 ). I 2 ≥50% or P<0.1 indicated significant heterogeneity, and the pooled results were evaluated by random-effects models; otherwise, fixed-effects model was conducted ( 28 ). Differences between the two groups (CA vs. non-CA; average daily treatment cost between different approval pathways) were determined by using the Mann-Whitney U test. Furthermore, this study evaluated the Pearson correlation coefficient between average daily cost and clinical benefits in different subgroups. Statistical analyses were conducted using Stata MP16.0. Figures were conducted using R (R Core Team, Vienna, Austria), version 4.1.1 (R package forestplot, version 3.1.3 and R package ggplot2, version 3.4.4). P<0.05 was considered to be statistically significant.

Results

Overview of approvals

From 2015 to 2023, a total of 120 new anticancer drugs (based on indication) were approved in China (details in Table S1 ), with 58 receiving CA, which constitutes a substantial 48.3% of the total approvals. The remaining 62 drugs were approved through non-CA pathways, including routine approval and priority review. Overall, there has been a consistent annual increase in the number of new anticancer drugs approved in China. Concurrently, the number of the CA drugs showed an annual increase during the first three years following the policy’s implementation. However, in 2023, there was a noticeable decrease ( Figure 1A ).

In addition, among the 58 CA anticancer drugs, the timeline to complete confirmatory trials set by the CDE [median 4 years, interquartile range (IQR) 3–5 years] were predominantly 4 years (31.0%) and 5 years (32.8%). A few drugs were required to complete their studies within 1–3 years, with the distribution being 6.9% for 1 year, 5.2% for 2 years, and 12.5% for 3 years. Larotrectinib was the only drug required to complete its studies within 6 years. As of the end of December 2024, according to the latest research and development (R&D) updates, 23 drugs (39.7%) had transitioned to routine review, 7 (12.1%) were at the supplementary application stage, and conditional studies for 26 drugs (44.8%) were ongoing. One drug (1.7%) was in other status, and another one (1.7%) had its drug registration certificate revoked due to failure to meet the primary endpoint in the confirmatory trial ( Figure 1B ).

In terms of the post-approval confirmatory clinical trials for drugs that have been converted to routine review, 23 drugs in our study involved a total of 26 post-approval confirmatory trials. Out of these, 14 drugs have completed their trials, while the remaining 9 drugs, despite still being underway, have completed the collection of major study results. Assessing the completion of these confirmatory clinical trials with the primary outcome completion time as the benchmark, 14 studies were completed within the timeline set by the CDE, while 7 were completed after it, with an average delay of 1.4 years (ranging from 0.5 to 3.7 years, calculated as 365 days per year), and 5 studies did not specify the time frame for completion or had not yet published the completion time of the main results.

Characteristics of pivotal clinical trials

Among the 120 drugs, a total of 170 pivotal clinical trials were conducted, with 82 being CA and 88 being non-CA. For the CA drugs, 16 conducted 2 pivotal clinical trials, and 42 conducted 1 pivotal clinical trials. The pivotal clinical trials for CA drugs were primarily phase II trials, accounting for 51.2%. In terms of trial types, 66 trials (80.5%) were single-arm. For the non-CA drugs, 14 conducted 2 pivotal clinical trials, and 48 conducted 1 pivotal clinical trial. The pivotal clinical trials were predominantly phase III trials, accounting for 70.5%. In terms of trial types, 67 trials (76.1%) were randomized controlled trials. There is no significant difference in the number of pivotal trials between CA and non-CA drugs.

Regarding the study endpoints, the majority of clinical trials employed a single primary study endpoint, while 9.4% (n=16) utilized multiple primary endpoints. In this study, 26 clinical trials (29.9%) adopted OS as primary or co-primary endpoint. Surrogate endpoints including PFS, ORR, complete response (CR), metastasis-free survival (MFS) and other were widely used as primary efficacy endpoints in pivotal trials, studies of non-CA drugs used PFS as primary study endpoint predominantly (CA vs. non-CA: 3.7% vs. 33.0%, P=0.11) while the studies of CA drugs mainly use ORR (CA vs. non-CA: 67.1% vs. 19.3%, P=0.01) ( Table 1 ).

Additionally, we performed a pooled analysis of ORR, PFS and OS. The pooled ORR was 59.6% [95% confidence interval (CI): 54.0–65.3%; I 2 =94.8%; Figure 2 , Table S2 ], while the pooled HR for OS was 0.75 (95% CI: 0.68–0.82; I 2 =51.8%; Figure 3 , Table S3 ) and the pooled HR for PFS was 0.50 (95% CI: 0.42–0.57; I 2 =84.6%; Figure 4 , Table S4 ). The median ORR, OS and PFS of CA drugs were 59.2%, 3.7 and 5.3 months, respectively. As for primary outcome, the pooled ORR for CA drugs was numerically higher than non-CA drugs (59.3% vs. 62.7%, P=0.51) ( Table 2 ) for single-arm clinical trials, as well in RCTs, no significant difference was found between CA drugs and non-CA drugs (pooled HR for OS: 0.66 vs. 0.82, P=0.03; pooled HR for PFS: 0.44 vs. 0.51, P=0.38) ( Table 2 ). As for secondary outcome, the CA drugs demonstrated a tendency of gains in OS and PFS compared to non-CA drugs (median PFS gains: 5.3 vs. 3.3 months, P=0.33; median OS gains: 3.7 vs. 2.5 months, P=0.32) ( Table 2 ).

Characteristics of confirmatory clinical trials

Among the 23 drugs that completed confirmatory trials, 26 post-approval confirmatory clinical trials were involved, of which 2 drugs were conducted with 1 China bridging study and 1 confirmatory study respectively, 1 drug was conducted with 2 confirmatory studies, and the remaining 21 drugs were conducted with 1 confirmatory study respectively. In terms of clinical trial design, single-arm designs were adopted in pre-approval pivotal clinical trials of 16 drugs (69.6%), while randomized controlled trials were adopted in the remaining 7 drugs (30.4%). In the post-approval confirmatory clinical trials, single-arm designs were utilized in the studies of 4 drugs (17.4%), mainly targeting leukemia and gastrointestinal stromal tumors; randomized controlled trials were utilized in the studies of the other 19 drugs (82.6%) to ascertain their therapeutic effects.

Clinical benefit of CA drugs in confirmatory trials versus pivotal trials

In terms of clinical benefit, the pre-market pivotal clinical trials primarily used the surrogate endpoint ORR as the primary endpoint (15/23, 65.2%), while confirmatory clinical trials mainly used PFS (16/23, 69.6%), and only 3 drugs (13.0%) used OS as the primary endpoint in confirmatory trials.

Regarding the concordance of endpoints between pivotal and confirmatory trials, 2 drugs (8.7%) used OS as the primary endpoint in both stages, 6 drugs (26.1%) used the same surrogate endpoints (ORR, CR/CRh, or PFS) in both stages, and 14 drugs (60.9%) used different surrogate endpoints in their confirmatory trials compared to their pivotal trials.

Usually, HR <0.75 and P<0.05 indicated clinically and statistically significant OS/PFS improvements. As for the result, the 2 drugs that used OS in both stages both achieved significant OS improvements in their confirmatory trials; all 6 drugs that used the same surrogate endpoints demonstrated clinical benefit in their confirmatory trials; and among the 15 drugs that used different endpoints, 11 (73.3%) showed both clinically and statistically significant benefit in their confirmatory trials, while results for the remaining 4 drugs (26.7%) were unavailable at the time of data cutoff. Details are presented in Table S5 and Figure 5 .

Average daily treatment cost

Of the 120 newly approved anticancer drugs that comprised 81 chemical drugs and 39 biological products in Figure 1A , 47 were domestic and 73 were imported in terms of origin. As for therapeutic areas, 86 targeted solid tumors, mainly non-small cell carcinomas, while 34 were for hematological cancers, with multiple myeloma being the most prevalent. It is worth noting that 86 of them were added to the NRDL, and 24 were first-in-class drugs. Drug origin and whether they were included in the NRDL indicated significant difference between CA and non-CA drugs. Compared with non-CA drugs, the CA drugs had a higher proportion of domestic production (CA vs. non-CA: 51.7% vs. 27.4%, P=0.006) and a lower rate of inclusion in the NRDL (62.1% vs. 80.6%, P=0.02) ( Table 3 ). Economically, 62.1% (36/58) of CA drugs entered NRDL [the average period from approved to enter NRDL is 1.5 years (IQR 0.5–3.4)], and 34.5% (20/58) entered NRDL within 1.5 years (some of them even in less than six months), with NRDL-included drugs showing greater price reductions (53.2% vs. 10.6%, P<0.001).

Overall, among 117 drugs [because of the differences in the dosage of drugs used for different indications in clinical use, the statistics of average daily costs are specifically broken down into indications. Based on the available data, the total number of drugs is 117 (6 drugs have no online prices and are not included in the statistics), of which multiple indications exist for septinib, orebactinib, and ibrutinib], the average reduction in daily cost from approval to the most recent data was 38.30%, with a median decrease of 38.41%. Specifically, drugs included in the NRDL (76/117) showed an average reduction of 53.22% and a median decrease of 61.44%; those not included in the NRDL (41/117) had an average reduction of 10.64% and a median decrease of 0.00%. Notably, daily cost reduction for drugs included in the NRDL was significantly greater than for those not included (53.22% vs. 10.64%, P<0.001). However, there were no significant difference in daily cost reduction between CA (average reduction: 36.15%; median decrease: 32.59%) and non-CA drugs (average reduction: 40.41%; median decrease: 45.20%) (P=0.42). The daily cost reduction between first-in-class (average reduction: 34.98%; median decrease: 16.74%) and non-first-in-class (average reduction: 39.11%; median decrease: 45.77%) drugs also indicated no significant difference (P=0.74) ( Table S6 ). Similarly, Table S7 also suggested no significant difference between CA and non-CA drugs in daily costs at the time of approval or in the latest data, as well in subgroup specific analysis of whether the drugs in the NRDL or not.

As for the correlation between drug daily costs at approval and latest with clinical benefits, there were no significant correlation between no matter daily treatment costs of approval or latest data and OS gains, PFS gains, or ORR. Subgroup analysis of approval pathway and in the NRDL or not also showed the same result, except for the analysis of drugs not in the NRDL, we found positive association between clinical benefit and daily cost of approval in PFS gains (r=0.958, P=0.01) ( Table 4 ).

Discussion

Since the implementation of the CA path, China’s drug CA system has played a positive role in the R&D and marketing of new drugs. From 2015, over half of the anticancer drugs approved have been conditionally approved, with a balanced proportion of domestic and imported drugs, significantly enhancing domestic innovation capabilities. Notably, the number of CA drugs have been declining since 2023, reflecting the gradual refinement and tightening of China’s CA policy by drug regulatory authorities. Tightened policy not only avoided a large number of homogeneous products on the market and encourage pharmaceutical companies to conduct more in-depth and innovative researches and development, but also helped to better manage risks associated with the inherent uncertainties of CA in terms of security. Admittedly, evaluating innovation capability cannot solely be measured by the number of new drugs approved; it also requires considering their specific clinical value ( 30 - 33 ). Therefore, this study aimed to compare the clinical values between CA and non-CA anticancer drugs and to explore the relationship between clinical benefit and the average daily cost of treatment of drugs. Our results showed that no significant differences in clinical benefits were observed between CA and non-CA drugs, as well no significant correlation between drug prices and clinical benefits was found, which supported that the standard to evaluate drug effectiveness have not been lowered by regulatory agencies due to CA process.

Between 2015 and 2023, 48.3% (n=58) of newly approved anticancer drugs received CA, reflecting the policy’s role in addressing urgent clinical needs. Importantly, CA and non-CA drugs demonstrated comparable clinical benefits, with no significant differences in pooled ORR (59.3% vs. 62.7%, P=0.51), PFS gains (5.3 vs. 3.3 months, P=0.33), or OS gains (3.7 vs. 2.5 months, P=0.32). Furthermore, the 23 (39.7%) drugs converted to routine review have shown continued benefits post-approval, in which the results of confirmatory clinical trials of all the 8 drugs used the same endpoints both in pre-approval pivotal trials and confirmatory trials improved numerically compared to pre-approval pivotal trials; except 4 drugs with unpublished confirmatory clinical results, the remaining 11 drugs utilized different surrogate endpoints showed both clinically and statistically significant benefit in their confirmatory trials. These findings underscore that during the drug review process, regulatory agencies have not lowered the criteria for evaluating drug effectiveness merely because the drugs are CA, the effectiveness requirements for CA drugs and non-CA are consistent and stringent.

It should be noted that the meta-analyses of ORR and PFS demonstrated substantial heterogeneity, as reflected by high I 2 values ( Table S5 ). This heterogeneity likely stems from the clinical diversity of the included studies, thus the pooled estimates should be interpreted with caution, and further studies are needed to explore potential sources of heterogeneity and clarify the robustness of the findings.

In addition, although there was no significant difference between CA and non-CA drugs in clinical benefits, firstly the CA pathway addressed unmet clinical needs by accelerating access to therapies for life-threatening conditions. Secondly, by shortening development timelines, CA enabled patient earlier to access treatments, which was particularly critical for populations with no existing treatments. Thirdly, the policy’s success is reflected in its support for domestic innovation (51.7% of CA drugs were domestically developed) and robust post-approval validation, with 95.7% (2 exhibited enhanced OS, 6 followed the same surrogate endpoints and increased, 14 demonstrated statistically significant efficacy using new surrogate endpoints) of CA drugs transitioned to routine review demonstrating sustained benefits in confirmatory trials. Furthermore, rapid NRDL inclusion (34.5% within 1.5 years) and significant price reductions also ensure affordability of drugs.

In the meanwhile, China’s CA system, though operational for less than a decade (since 2015), has shown early progress in aligning pre-approval and confirmatory trial outcomes. Notably, 39.7% (n=23) of CA drugs transitioned to routine review and completed confirmatory trials within regulatory timelines (predominantly 4–5 years), with only 1.7% (n=1) withdrawn. These early results indicate that China’s CA framework has demonstrated a commitment to timely evidence generation and regulatory accountability. However, given the relatively short operational history of the system, its long-term sustainability and effectiveness require continued monitoring and evaluation as more drugs complete the full post-approval cycle.

In terms of the type of confirmatory clinical benefit, the primary study endpoint of 20 drugs were surrogate endpoints. For specific types of tumors, the surrogate endpoints can verify their benefits ( 34 , 35 ), while on the contrast, multiple studies in Europe and the U.S. have shown that drugs approved by the FDA/European Medicines Agency (EMA) based on surrogate endpoints are not always able to demonstrate a positive impact on OS in subsequent confirmatory clinical trials ( 19 , 36 - 41 ). Moreover, the validity of surrogate endpoints and the magnitude of observed effect sizes may also vary across treatment lines, and trial designs (single-arm vs. RCT), which should be considered when interpreting cross-trial comparisons of clinical benefit between CA and non-CA drugs. Therefore, selecting scientifically validated surrogate endpoints is crucial. China can draw on FDA’s experience to develop and regularly update a list of endorsed surrogate endpoints, guiding developers in making more precise clinical trial choices. Additionally, as further evidence accumulates post-approval, the assessment of clinical value may evolve, underscoring the importance of robust post-marketing evaluation for CA drugs. Regulatory agencies must closely monitor the progress of post-marketing studies to ensure timely submission of required data and evidence ( 42 - 44 ).

As for treatment cost, new anticancer drugs, particularly those included in the NRDL, have experienced substantial price reductions post-approval, with some experiencing a maximum decrease of up to 95.77%. This trend is attributed to the national medical insurance negotiation policy of China (established in 2017), which facilitates the annual inclusion of innovative drugs in the catalog of medicines covered by national medical insurance system through the establishment of a negotiation floor price (termed the “intentional purchase price”). This price is determined by evaluating drug innovation, cost-effectiveness, and financial sustainability. However, this study revealed no significant difference in daily cost reduction between CA (average reduction: 36.15%; median decrease: 32.59%) and non-CA drugs (average reduction: 40.41%; median decrease: 45.20%) (P=0.42), together with no significant correlation was found between clinical benefits and drug prices, both at the time of approval and subsequent to approval, consistent with the research of Zhang [2022] ( 45 ) and Yuan [2023] ( 46 ), suggesting poor alignment between drug value and price. Not only the opacity of negotiation processes ( 47 ) but also the reliance on incompletely validated efficacy data at approval complicate value assessment, especially for CA drugs, whose benefits often incompletely validated at approval and require post-approval confirmation. Therefore, future pricing strategies for innovative drugs requires to be further improved, for example, focus more on potential clinical benefits and prioritize CA drugs for their innovation value and higher R&D investments.

It should be noted that our assessment of clinical benefit relied on conventional efficacy endpoints (ORR, OS, and PFS) extracted from pivotal clinical trials, rather than validated benefit assessment frameworks. Previous studies have employed tools such as the ESMO-MCBS ( 48 ) or ASCO Value Framework ( 49 ) to provide more standardized and granular evaluations of the magnitude of clinical benefit for approved anticancer drugs. These frameworks incorporate additional dimensions such as quality of life, toxicity profiles, and the magnitude of survival gains relative to existing therapies, which were not captured in our analysis. Future studies applying these validated frameworks to CA drugs in China would provide a more comprehensive and internationally comparable assessment of their clinical value.

There are several limitations in this study. First, the comparison between CA and non-CA drugs did not adjust for key clinical development factors that may influence both clinical benefit and pricing, including cancer type, treatment line, biomarker-defined patient populations, and first-in-class status. In addition, as shown in Table 1 , CA and non-CA drugs differ fundamentally in the phase and design of their supporting pivotal trials—CA drugs are predominantly approved based on Phase II single-arm trials, whereas non-CA drugs are largely supported by Phase III RCTs. These structural differences in trial design may influence observed effect sizes for ORR, PFS, and OS, and thus represent potential confounders when directly comparing clinical benefits between the two groups. Although these differences are inherent to the regulatory pathways rather than a result of selection bias, stratified analyses by trial phase or trial design would help assess the robustness of the findings. Due to the relatively small sample size and heterogeneity of indications, meaningful stratified analyses were not feasible. Future studies with larger datasets should explore subgroup and sensitivity analyses to better isolate the effect of the approval pathway from these confounding factors. Second, confirmatory trials were not systematically identified or included, as many CA drugs had not yet completed their confirmatory trials or published results in peer-reviewed journals at the time of data cutoff. Our analysis therefore reflects the clinical evidence available at and after approval rather than the full post-approval evidence trajectory. Future research should systematically evaluate confirmatory trial outcomes as more data become available. Third, the price data came from a public database, not a confidential health technology assessment report. And due to the confidentiality of the national medical insurance negotiation process, it’s difficult to determine to what extent the clinical value of drugs in the CA pathway has been considered for inclusion in the medical insurance negotiation.

Conclusions

In conclusion, the CA process is pivotal in bridging the gap in clinical drug availability and ensuring patient access to medications and that regulatory agencies have not lowered the drug effectiveness evaluation standard in CA process. In addition, the confirmatory trials outcomes of CA drugs aligned with that of pre-approval trials in this study due to China’s CA system, yet the NMPA should further balance unmet clinical needs with clinical value, offering guidance for R&D of CA drugs to ensure that companies conduct necessary post-approval confirmatory clinical trials to verify clinical benefits. Additionally, the NHSA should develop improved transparent, open, scientific, and flexible medical insurance negotiation policies to enhance the alignment of clinical value with pricing.

Supplementary

The article’s supplementary files as

Acknowledgments

None.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Footnotes

Reporting Checklist: The authors have completed the CHEERS reporting checklist. Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2026-1-0038/rc

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2026-1-0038/coif ). The authors have no conflicts of interest to declare.

References

1. Mehta GU, de Claro RA, Pazdur R. Accelerated Approval Is Not Conditional Approval: Insights From International Expedited Approval Programs. JAMA Oncol 2022;8:335-6. 10.1001/jamaoncol.2021.6854

2. Copland M. Treatment of blast phase chronic myeloid leukaemia: A rare and challenging entity. Br J Haematol 2022;199:665-78. 10.1111/bjh.18370

3. Madhusoodanan J. How a controversial US drug policy could be harming cancer patients worldwide. Nature 2023;620:264-7.

4. Su L, Liu S, Li G, et al. Trends and Characteristics of New Drug Approvals in China, 2011-2021. Ther Innov Regul Sci 2023;57:343-51. 10.1007/s43441-022-00472-3

5. Wang F, Ruan DY, Xu RH. Challenges and opportunities in oncology drug development and clinical research in China. Cell 2024;187:1578-83. 10.1016/j.cell.2024.02.040

6. Li G, Liu Y, Xie C, et al. Characteristics of expedited programmes for cancer drug approval in China. Nat Rev Drug Discov 2021;20:416. 10.1038/d41573-021-00080-0

7. Luo X, Qian F, Yang L, et al. Assessment of the breakthrough-therapy-designated drugs granted in China: A pooled analysis 2020-2022. Drug Discov Today 2022;27:103370. 10.1016/j.drudis.2022.103370

8. Zhang P, Zhang M, Ma R, et al. Characteristics of indications, clinical trial evidence, clinical benefits and the costs of price-negotiated multi-indication drugs for solid tumours in China. J Glob Health 2025;15:04121. 10.7189/jogh.15.04121

9. Zhu T, Zhong J, Huang Y. Cancer Drugs Approved Based on Surrogate Endpoint: A Retrospective Observational Study in the United States and China. Cancer Med 2025;14:e70864. 10.1002/cam4.70864

10. Tang L, Song Y, Zhang H, et al. Anti-tumor drug supervision in China from 2010 to 2024: the evolution and prospect of drug review standards. Cancer Commun (Lond) 2025;45:63-7. 10.1002/cac2.12630

11. Guan H, Shi Y, Song J, et al. Impact of competition on reimbursement decisions for cancer drugs in China: an observational study. Lancet Reg Health West Pac 2024;50:101157. 10.1016/j.lanwpc.2024.101157

12. Tibau A, Romano A, Liu ITT, et al. Assessing outcomes emerging after conversion to regular approval for cancer drug indications granted accelerated approval, 1992-2021. J Natl Cancer Inst 2025;117:2103-11. 10.1093/jnci/djaf195

13. Patel SR, Ramachandran M, Ai A, et al. Redefining Available Therapy in Oncology Accelerated Approval Decisions. J Clin Oncol 2025;43:629-32. 10.1200/JCO.24.00892

14. Tibau A, Hwang TJ, Romano A, et al. Factors in Time to Full Approval or Withdrawal for Anticancer Medicines Granted Accelerated Approval by the FDA. JAMA Netw Open 2025;8:e252026. 10.1001/jamanetworkopen.2025.2026

15. Gyawali B, Rome BN, Kesselheim AS. Regulatory and clinical consequences of negative confirmatory trials of accelerated approval cancer drugs: retrospective observational study. BMJ 2021;374:n1959. 10.1136/bmj.n1959

16. Tibau A, Cliff ERS, Romano A, et al. Predictors of withdrawal of anticancer drug indications granted accelerated approval: a retrospective cohort study. EClinicalMedicine 2025;84:103088. 10.1016/j.eclinm.2025.103088

17. Downing NS, Shah ND, Aminawung JA, et al. Postmarket Safety Events Among Novel Therapeutics Approved by the US Food and Drug Administration Between 2001 and 2010. JAMA 2017;317:1854-63. 10.1001/jama.2017.5150

18. Zhang L, Zhang Y, Li H, et al. Improvements in the health-related quality-of-life benefit of cancer drugs approved in China, 2005-2020: an observational study. Lancet Reg Health West Pac 2024;46:101077. 10.1016/j.lanwpc.2024.101077

19. Luo X, Du X, Huang L, et al. Evidence of pre-approval clinical trial supporting the granted conditional approval for novel cancer drugs in China between 2015 and 2022. EClinicalMedicine 2023;63:102177. 10.1016/j.eclinm.2023.102177

20. National Medical Products Administration . Available online: https://www.nmpa.gov.cn/

21. National Medical Products Administration. Listed drug information. Available online: https://www.cde.org.cn/main/xxgk/listpage/b40868b5e21c038a6aa8b4319d21b07d

22. Pharmacodia. Available online: https://www.pharmacodia.com/

23. Eliancloud. Available online: https://yp.eliancloud.cn/

24. Ratain MJ. Body-surface area as a basis for dosing of anticancer agents: science, myth, or habit? J Clin Oncol 1998;16:2297-8. 10.1200/JCO.1998.16.7.2297

25. World Health Organization Collaborating Centre for Drug Statistics Methodology. Definition and general considerations of the defined daily dose 2018. Available online: https://atcddd.fhi.no/ddd/definition_and_general_considera/

26. Vokinger KN, Hwang TJ, Grischott T, et al. Prices and clinical benefit of cancer drugs in the USA and Europe: a cost-benefit analysis. Lancet Oncol 2020;21:664-70. 10.1016/S1470-2045(20)30139-X

27. Higgins JP, Thompson SG, Deeks JJ, et al. Measuring inconsistency in meta-analyses. BMJ 2003;327:557-60. 10.1136/bmj.327.7414.557

28. Sedgwick P. Meta-analyses: what is heterogeneity? BMJ 2015;350:h1435. 10.1136/bmj.h1435

29. A Study of DCC-2618 (Ripretinib) In Patients With With Advanced Gastrointestinal Stromal Tumors (GIST). Available online: https://clinicaltrials.gov/study/NCT04282980?term=NCT04282980&rank=1&tab=results

30. Kesselheim AS, Wang B, Avorn J. Defining "innovativeness" in drug development: a systematic review. Clin Pharmacol Ther 2013;94:336-48. 10.1038/clpt.2013.115

31. Zhang Y, Wagner AK, Guan X. Newly approved cancer drugs in China - innovation and clinical benefit. Nat Rev Clin Oncol 2023;20:135-6. 10.1038/s41571-023-00728-3

32. Hakariya H, Ozaki A, Hashimoto T, et al. Added Therapeutic Benefits of Top-Selling Drugs in Japan: A Cross-Sectional Study Using Health Technology Assessment. Clin Transl Sci 2025;18:e70243. 10.1111/cts.70243

33. Pan J, Wei X, Lu H, et al. List prices and clinical value of anticancer drugs in China, Japan, and South Korea: a retrospective comparative study. Lancet Reg Health West Pac 2024;47:101088. 10.1016/j.lanwpc.2024.101088

34. McKee AE, Farrell AT, Pazdur R, et al. The role of the U.S. Food and Drug Administration review process: clinical trial endpoints in oncology. Oncologist 2010;15 Suppl 1:13-8.

35. Blasco A, López-Tarruella S, Urruticoechea A, et al. Intermediate endpoints as surrogates for long-term outcomes in breast cancer adjuvant chemotherapy trials. Oncologist 2025;30:oyaf263. 10.1093/oncolo/oyaf263

36. Haslam A, Hey SP, Gill J, et al. A systematic review of trial-level meta-analyses measuring the strength of association between surrogate end-points and overall survival in oncology. Eur J Cancer 2019;106:196-211. 10.1016/j.ejca.2018.11.012

37. Chen EY, Haslam A, Prasad V. FDA Acceptance of Surrogate End Points for Cancer Drug Approval: 1992-2019. JAMA Intern Med 2020;180:912-4. 10.1001/jamainternmed.2020.1097

38. Dawoud D, Naci H, Ciani O, et al. Raising the bar for using surrogate endpoints in drug regulation and health technology assessment. BMJ 2021;374: n2191. 10.1136/bmj.n2191

39. Gyawali B, Hey SP, Kesselheim AS. Evaluating the evidence behind the surrogate measures included in the FDA's table of surrogate endpoints as supporting approval of cancer drugs. EClinicalMedicine 2020;21:100332. 10.1016/j.eclinm.2020.100332

40. Mooghali M, Mitchell AP, Skydel JJ, et al. Characterization of accelerated approval status, trial endpoints and results, and recommendations in guidelines for oncology drug treatments from the National Comprehensive Cancer Network: cross sectional study. BMJ Med 2024;3:e000802. 10.1136/bmjmed-2023-000802

41. Shahzad M, Naci H, Esselen KM, et al. Regulatory histories of recently withdrawn ovarian cancer treatment indications of 3 PARP inhibitors in the US and Europe: lessons for the accelerated approval pathway. J Pharm Policy Pract 2024;17:2351003. 10.1080/20523211.2024.2351003

42. Benjamin DJ, Lythgoe MP. Modernising the US FDA's Accelerated Approval pathway. Lancet Oncol 2023;24:203-5. 10.1016/S1470-2045(23)00020-7

43. Fashoyin-Aje LA, Mehta GU, Beaver JA, et al. The On- and Off-Ramps of Oncology Accelerated Approval. N Engl J Med 2022;387:1439-42. 10.1056/NEJMp2208954

44. Mooghali M, Wallach JD, Ross JS, et al. Premarket Pivotal Trial End Points and Postmarketing Requirements for FDA Breakthrough Therapies. JAMA Netw Open 2024;7:e2430486. 10.1001/jamanetworkopen.2024.30486

45. Zhang Y, Wei Y, Li H, et al. Prices and Clinical Benefit of National Price-Negotiated Anticancer Medicines in China. Pharmacoeconomics 2022;40:715-24. 10.1007/s40273-022-01161-7

46. Yuan J, Li M, Lu ZK. National value-based pricing negotiation for oncology drugs - lessons from China. Nat Rev Clin Oncol 2023;20:501-2. 10.1038/s41571-023-00769-8

47. Liu GG, Wu J, He X, et al. Policy Updates on Access to and Affordability of Innovative Medicines in China. Value Health Reg Issues 2022;30:59-66. 10.1016/j.vhri.2021.12.003

48. Cherny N, Oosting S, Dafni U, et al. ESMO-Magnitude of Clinical Benefit Scale version 2.0 (ESMO-MCBS v2.0). Annals of Oncology 2025;36:866-908.

49. Schnipper LE, Davidson NE, Wollins DS, et al. Updating the American Society of Clinical Oncology Value Framework: Revisions and Reflections in Response to Comments Received. J Clin Oncol 2016;34:2925-34. 10.1200/JCO.2016.68.2518