Consistency of primary outcomes in evidence synthesis researches on drug-eluting stents for coronary artery disease Authors: Xiaoqing Lin, Songsong Tan, Yuezi Huang, Jiaojiao Zhao, Jiankun Yao, Die Han, Renlan Luo, Huan Wu, Shuimei Sun, Junjie Lan, Rui Zhang, Huaye Zhao, Linfang Hu, Jiaxue Wang, Wenyi Zheng, Rui He, Jiaxing Zhang Published: 2025 ABSTRACT AbstractBackgroundAs the number of systematic reviews (SRs) and meta-analyses (MAs) evaluating drug-eluting stents (DES) for coronary artery disease (CAD) continues to grow, the need for standardized primary outcomes has become increasingly important. This study aimed to examine the specification and selection of primary outcomes in published SRs/MAs on DES for CAD and to identify factors associated with their reporting.MethodsWe conducted a cross-sectional analysis of SRs/MAs on DES for CAD. They were retrieved from English-language databases (PubMed, Embase, Cochrane Library) and Chinese-language databases (CNKI, Wanfang, VIP) from the beginning of their establishment to May 3, 2025. We assessed whether primary outcomes were explicitly specified and employed multivariable logistic regression to identify factors associated with such specification. For frequently reported outcomes, we compared how often they were designated as primary outcomes. In SRs/MAs that clearly stated primary outcomes, we categorized them using the Core Outcome Measures in Effectiveness Trials (COMET) framework and calculated the Corrected Covered Area (CCA) to assess outcome overlap.ResultsA total of 271 SRs/MAs from 29 countries or regions were included. Only 188 (188/271, 69.37%) specified primary outcomes. Specification was significantly associated with prospective protocol registration (adjusted odds ratio = 2.69, 95% confidence interval: 1.11–6.54, P = 0.03). We identified 18 distinct primary outcomes among these reviews, with a low overlap (Corrected Covered Area = 0.0544). The most frequently reported outcomes were specified as primary outcomes in just 19.56% (53/271) of studies, while the most designated primary outcomes were not reported in 43.54% (118/271) of reviews. According to the COMET framework, coverage of life impact, resource use, and other non-physiological domains was limited.ConclusionThe inconsistent specification and limited standardization of primary outcomes in SRs/MAs of DES for CAD may undermine evidence quality. The pathway from “unspecified primary outcomes” to “lack of standardization” to an “incomplete core outcome set” suggests a need for urgent methodological improvement in outcome reporting.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12874-025-02725-9.Keywords: Primary outcome, Coronary artery disease, Systematic review, Core outcome set, Evidence quality Background Cardiovascular diseases (CVDs) have become the leading global public health concern [1, 2]. Driven by population aging and shifts in lifestyle, the burden of CVDs is rising sharply. Between 2025 and 2050, the global prevalence of CVDs is projected to increase by 90.0%, with crude mortality rising by 73.4% and disability-adjusted life years (DALYs) by 54.7%. Cardiovascular deaths are expected to reach 35.6 million annually by 2050 [3]. As a major subtype of CVDs, coronary artery disease (CAD) contributes substantially to this escalating burden. In the management of CAD, drug-eluting stents (DES) have become the cornerstone of percutaneous coronary intervention, largely replacing bare-metal stents and balloon angioplasty due to their superior efficacy and safety profile. DES are now used in nearly all interventional contexts [4]. Given that outcomes (e.g. stent thrombosis, target lesion revascularization, target vessel revascularization, target lesion failure, late lumen loss, repeat revascularization, restenosis, target vessel failure, and device-oriented composite endpoints) in device-related research are more complicated than those in pharmacological interventions research, inconsistencies in outcome reporting are problematic in device-related research compared to pharmacological interventions. Despite widespread use, systematic reviews (SRs) and meta-analyses (MAs) evaluating DES for CAD often exhibit inconsistencies in how primary outcomes are specified. Some studies list multiple primary outcomes [5–8], which contradicts the fundamental methodological principle that a primary outcome should reflect the central research objective. Even among studies that designate a single primary outcome, there is considerable variability in outcome selection, ranging from target vessel revascularization (TVR) and stent thrombosis (ST) to major adverse cardiovascular events (MACE) and all-cause mortality [9–14]. This heterogeneity undermines the comparability of studies, compromises the validity of evidence synthesis, and hinders the formation of clear, actionable clinical conclusions. The lack of standardized primary outcomes not only complicates meta-analytic integration but also impairs the development of evidence-based clinical guidelines. When guidelines are based on inconsistent endpoints, they risk offering recommendations that are not adequately supported by robust or comparable evidence, ultimately affecting the quality of patient care. International initiatives, such as the Core Outcome Measures in Effectiveness Trials (COMET) Core Outcome Sets (COS) [15] and the Consolidated Standards of Reporting Trials (CONSORT) Statement [16] aim to standardize outcome reporting in clinical research. However, to date, no study has systematically assessed the practices, patterns, and determinants of primary outcome specification in SRs/MAs focused on DES for CAD. To address this gap, we conducted a cross-sectional survey of existing SRs/MAs on DES for CAD. Our objectives were to characterize how primary outcomes are reported, identify factors associated with their specification, and provide a foundation for developing a standardized COS in this domain. Methods We followed the STROBE (strengthening the reporting of observational studies in epidemiology) statement to conduct and report the present study (Supplementary Table S1). Search strategy A comprehensive literature search was conducted in three English-language databases (PubMed, Embase, and the Cochrane Library) and three Chinese-language databases (CNKI, Wanfang, and VIP) from their inception to May 3, 2025. Detailed search strategies for each database are provided in Supplementary Table S2. Eligibility criteria We included published systematic reviews (SRs) and/or meta-analyses (MAs) that met the following criteria: the study population consisted of patients diagnosed with coronary artery disease (CAD), without restriction on age or sex; the intervention group received drug-eluting stents (DES), and the control group received any alternative therapy, including other DES, bare-metal stents (BMS), drug-eluting balloons (DEB), drug-coated balloons (DCB), coronary artery bypass grafting (CABG), or pharmacological treatment; and the article was published in either English or Chinese. Studies were excluded if they were duplicates or if the full text was unavailable. Study selection and data extraction Two reviewers independently screened titles and abstracts for relevance based on the eligibility criteria, followed by full-text assessment for final inclusion. After completing standardized training, the same two reviewers independently extracted data from each included SR/MA. Extracted data included general information (first author, year of publication, PROSPERO registration number, population characteristics, interventions, comparators, and both primary and secondary outcomes) as well as methodological details (types of included studies, tools used for risk of bias assessment, methods for evaluating the certainty of evidence, and other relevant variables). Any inconsistencies during screening and data extraction were resolved by consensus or through adjudication by a third party. All discrepancies identified between the two reviewers were resolved by consensus, with the exception of a few persistent disagreements on subjective points (e.g., classification of ambiguous outcomes), which were referred to a third reviewer for a final decision. Statistical analysis Inter-reviewer agreement for study selection was evaluated using Cohen’s kappa statistic (K), which ranges from 0 to 1 [17, 18]. Statistical analyses were conducted using R software (version 4.3.1), with a two-sided significance level set at P < 0.05. Categorical variables were reported as frequencies and proportions, while continuous variables were presented as mean ± standard deviation for normally distributed data or median with interquartile range (IQR) for non-normally distributed data. Differences in categorical variables were assessed using chi-square or Fisher’s exact tests. For continuous variables, t-tests or z-tests were used when data were normally distributed, and the Wilcoxon rank-sum test was used for non-normally distributed or ordinal data. To explore the factors associated with whether a SR/MA specified primary outcomes, we applied multivariable logistic regression models. The potential explanatory variables included year of publication (before vs. after 2011), protocol registration status (yes vs. no), geographic origin of the research group (Chinese vs. non-Chinese), inclusion of multiple study types (> 1 vs. 1), performance of risk of bias assessment (yes vs. no), and use of the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach to evaluate certainty of evidence (yes vs. no). We summarized the frequency and proportion of outcomes reported across all included SRs/MAs, with particular attention to how often each outcome was designated as a primary outcome. A co-occurrence network map was generated to visualize clustering patterns among the primary outcomes. To quantify the extent of overlap in primary-outcome selection across studies, we calculated the Corrected Covered Area (CCA) as (N − r)/[(r × c) − r], where N is the total number of primary-study citations (including duplicates), r is the number of unique studies, and c is the number of included reviews [19]. Following Pieper et al.‘s [20] benchmarks, CCA < 5% indicates slight overlap, 5–15% moderate overlap, and ≥ 15% high overlap. Finally, we categorized the specified primary outcomes into five domains—death, physiological or clinical, life impact, resource use, and adverse events—based on the taxonomy provided by the COMET Initiative [15, 21]. Results Search results and characteristics of included studies A total of 2,763 records were initially identified through searches of six databases. The inter-reviewer agreement for study selection was high, with a Cohen’s kappa statistic of 0.91, indicating almost perfect consistency. After removing duplicates and screening based on eligibility criteria (Fig. 1), 271 SRs/MAs studies were included in the final analysis (Supplementary Table S3). Flow diagram of study selection process for this study. SR: systematic review; MA: Meta-Analyses; DES: Drug-eluting Stent The studies included were published between 2005 and 2025, with a marked increase in publications after 2011, accounting for 88.19% (239/271) of the total. The studies were authored by researchers from 29 different countries or regions, with 34.32% (93/271) originating from China. Only a small proportion of studies (46/271, 16.97%) had prospectively registered protocols. Regarding clinical focus, the majority of studies (197/271, 72.69%) targeted general CAD, while others addressed specific patient subgroups, including CAD with diabetes (21/271, 7.75%), CAD with renal disease (10/271, 3.69%), high bleeding risk (2/271, 0.74%), hypertension (1/271, 0.37%), and acute coronary syndrome (ACS; 40/271, 14.76%). In terms of study designs, 205 studies (205/271, 75.65%) focused on a single type of primary study. Among these, randomized controlled trials (RCTs) were by far the most prevalent, included in 192 reviews (192/205, 93.66%), followed by cohort studies (7/205, 3.41%) and other observational studies (6/205, 2.93%). The remaining 66 SRs/MAs (66/271, 24.35%) incorporated multiple types of studies. Of these, 43 reviews (43/66, 65.15%) included both RCTs and cohort studies (CS), 15 (15/66, 22.73%) included RCTs and other observational studies (OS), 6 (6/66, 9.09%) included RCTs and non-randomized controlled trials (NRCTs), and 2 (2/66, 3.03%) combined RCTs, CS, and case-control studies. Out of the 271 SRs/MAs, 182 (182/271, 67.16%) evaluated the methodological quality of their included studies. Among these, 11 (11/182, 6.04%) reviews that included both RCTs and OS applied the Cochrane Risk of Bias tool exclusively to the RCTs. Only 17 studies (17/271, 6.27%) used the GRADE system to assess the certainty of evidence. Factors associated with primary outcome specification Among the 271 SRs/MAs, 188 (188/271, 69.37%) clearly specified primary outcomes. Univariate analysis revealed that the presence of a prospectively registered protocol was significantly associated with primary outcome specification (odds ratio [OR] = 2.84, 95% confidence interval [CI]: 1.21–6.65, P = 0.02, Table. 1), as was the assessment of methodological quality of included studies (OR = 1.81, 95% CI: 1.06–3.10, P = 0.03). However, in multivariable logistic regression adjusting for other potential factors, including year of publication, origin of research group (Chinese vs. non-Chinese), number of study types included, quality assessment, and use of GRADE, only prospective protocol registration remained statistically significant (adjusted OR = 2.69, 95% CI: 1.11–6.54, P = 0.03). Table 1Univariate and multivariate analyses of factors influencing the setting of primary outcomeVariablesWith primary outcome (n = 188)Without primary outcome(n = 83)Univariate analysisMultivariate analysisOR95%CIPAOR95%CIPPublished time after/before 2011168/2071/121.42[0.66, 3.06]0.371.01[0.44, 2.33]0.97Registered protocol with/without protocols39/1497/762.84[1.21, 6.65]0.02*2.69[1.11, 6.54]0.03*Research Group Non-Chinese/Chinese125/6353/301.12[0.65, 1.93]0.671.20[0.67, 2.17]0.54The types of included studies >one/one44/14423/600.80[0.44, 1.43]0.450.85[0.46, 1.58]0.61Assess the quality of included studies Yes/No134/5448/351.81[1.06, 3.10]0.03*1.75[0.97, 3.18]0.07Assess the certainty of evidence using GRADE Yes/No12/1765/781.06[0.36, 3.12]0.910.67[0.21, 2.13]0.50OR odds ratio, CI confidence interval, AOR adjusted odds ratio *: P < 0.05 Univariate and multivariate analyses of factors influencing the setting of primary outcome OR odds ratio, CI confidence interval, AOR adjusted odds ratio *: P < 0.05 Reporting patterns and primary outcome selection Across all 271 included SRs/MAs, a total of 19 distinct outcomes were reported and most of them are qualitative except for late lumen loss (LLL) and quality of life. The most frequently reported outcome was myocardial infarction (MI), appearing in 243 studies (243/271, 89.67%). This was followed by all-cause mortality (187/271, 69.00%), stent thrombosis (ST; 156/271, 57.56%), target lesion revascularization (TLR; 153/271, 56.46%), and target vessel revascularization (TVR) or major adverse cardiovascular events (MACE), each reported in 125 studies (125/271, 46.13%), as shown in Fig. 2. Frequency distribution of primary vs. All outcome measures in included studies. SRs: systematic reviews; MA: Meta-Analyses Among the 188 SRs/MAs that specified primary outcomes, 18 different outcome types were designated as primary, resulting in a low degree of overlap, with a Corrected Covered Area (CCA) of 0.0544 (Figure S1). Notably, MI, despite being the most reported outcome, was specified as the primary outcome in only 19.56% (53/271) of the reviews. In contrast, TLR, which was the most frequently used primary outcome, was entirely unreported in 118 SRs/MAs (118/271, 43.54%). Approximately one-third of the SRs/MAs that specified primary outcomes (56/188, 29.79%) listed more than two primary outcomes (Fig. 3). Among the 98 SRs/MAs that identified only one primary outcome, 10 unique endpoints were used. Of these, MACE was the most frequently selected, accounting for 26.53% (26/98) of single-outcome studies. Frequency distribution of primary outcomes and corresponding systematic reviews/meta-analyses. MACE: Major adverse cardiovascular events; TLR: Target lesion revascularization; TVR: Target vessel revascularization; TLF: Target lesion failure; ST: Stent thrombosis; MACCE: Major adverse cardiac and cerebrovascular events; MI: Myocardial infarction; LLL: Late lumen loss; TVF: Target vessel failure Classification of primary outcomes under the COMET framework Following the COMET Initiative framework, the 18 primary outcomes identified across SRs/MAs were classified into five domains: death, physiological or clinical, life impact, resource use, and adverse events. Most of the primary outcomes (13/18, 72.22%) fell within the adverse events domain (Fig. 4). The other three (physiological or clinical, life impact, resource use) domains were represented by only one outcome each, indicating a substantial underrepresentation of non-adverse event domains in outcome selection across the included reviews. Classification and study frequencies of outcome measures in drug-eluting stent interventions for coronary artery disease. LLL: Late lumen loss; MI: Myocardial infarction; MACE: Major adverse cardiovascular events; MACCE: Major adverse cardiac and cerebrovascular events; ST: Stent thrombosis; TLF: Target lesion failure; TVF: Target vessel failure; DOCE: Device-oriented composite endpoints; POCE: Patient-oriented composite endpoints; TVR: Target vessel revascularization; TLR: Target lesion revascularization Discussion Main findings and interpretations This study investigated 271 systematic reviews and meta-analyses (SRs/MAs) evaluating the efficacy and safety of drug-eluting stents (DES) for the treatment of coronary artery disease (CAD), covering research conducted in 29 countries or regions. Our findings reveal substantial inconsistencies in the specification of standardized primary outcomes and highlight a notable gap in the development of a core outcome set (COS) in this domain. Approximately one-third of SRs/MAs failed to define a primary outcome, a shortcoming that could potentially be addressed through prospective protocol registration. Moreover, the mismatch between frequently reported outcomes and those most designated as primary outcomes indicates an absence of consensus on outcome prioritization. For instance, the most reported outcomes were specified as primary outcomes in fewer than 20% of the studies, while the most used primary outcomes were not reported at all in over 40% of the reviews. Among the 188 SRs/MAs that did specify primary outcomes, 18 distinct outcomes were identified, with minimal overlap, highlighting remarkable heterogeneity in endpoint selection. This variability was particularly pronounced in SRs/MAs that specified only a single primary outcome. Within the COMET framework, outcomes related to physiological or clinical impact, life impact, and resource use were underrepresented, indicating a narrow focus on adverse event endpoints. These patterns suggest that inconsistent practice may limit comparability of evidence, which we conceptualize as progressing from an absence of specified primary outcomes to a lack of standardized outcome selection, ultimately leading to an incomplete COS (Table 1). A clearly defined primary outcome is a cornerstone of high-quality research. As the outcome deemed most critical to the research objective, the primary outcome minimizes the risk of type I errors arising from multiple comparisons and supports appropriate sample size calculations to prevent type II errors [22, 23]. Given their status at the top of the evidence hierarchy, SRs/MAs are expected to prespecify their primary outcome before study initiation. However, our study revealed that a considerable proportion of SRs/MAs on DES for CAD failed to meet this standard, increasing susceptibility to selective outcome reporting. Importantly, we found that protocol registration significantly increased the likelihood of specifying a primary outcome. Registries such as PROSPERO require declaration of the primary outcome at the time of registration, thereby promoting methodological rigor from the design phase onward [24]. We also observed a trend suggesting that SRs/MAs assessing the methodological quality of included studies were more likely to prespecify primary outcomes. This may be attributable to the fact that proper evaluation of reporting bias in included studies depends on identifying prespecified primary outcomes. Widely used quality assessment tools such as the Cochrane Risk of Bias tool for randomized controlled trials (RCTs) and the Newcastle–Ottawa Scale (NOS) for observational studies include items that evaluate the appropriateness of predefined outcome measures. The limited reporting of commonly used primary outcomes in SRs/MAs suggests a lack of universally accepted core outcomes for CAD. Even among those studies that defined a primary outcome, endpoint selection was inconsistent and scattered, which compromises the comparability and synthesis of findings across reviews [25–27]. Establishing a COS—defined as the minimum set of outcomes that should be measured and reported in clinical studies of a specific health condition—could help resolve these issues. Applying the COMET classification system, we found that nearly 80% of primary outcomes were categorized as adverse event-related endpoints, including both single outcomes and composite outcomes. While single endpoints provide clear, direct assessments of specific device performance, patients with CAD often experience multiple adverse events concurrently [28]. Consequently, focusing on single endpoints may inflate the likelihood of false-positive findings due to repeated testing. Composite endpoints mitigate this risk by integrating multiple related outcomes, potentially offering a more clinically meaningful assessment of therapeutic benefit. Of the composite outcomes, MACCE stands out for its inclusion of both cardiovascular and cerebrovascular components, offering a comprehensive evaluation of adverse events. Despite being specified as the primary outcome in only eight SRs/MAs, we suggest that MACCE warrants consideration as a key standardized outcome. Beyond adverse events, outcomes related to life impact and resource use remain largely underreported in the current literature, despite their relevance for clinical decision-making. For example, the International Consortium for Health Outcomes Measurement (ICHOM) recommends a standard set of CAD outcomes that includes patient-reported measures such as the Seattle Angina Questionnaire (SAQ-7), Rose Dyspnea Scale, and Patient Health Questionnaire (PHQ-2), as well as metrics related to hospital admissions and cardiovascular procedures [29]. Similarly, the Meetbaar Beter initiative has implemented comprehensive patient-centered outcomes for CAD across 21 Dutch hospitals, including quality of life assessments using the SF-36 questionnaire [30]. The European Society of Cardiology’s Working Group on Thrombosis has also emphasized bleeding as a clinically significant outcome in acute coronary syndrome (ACS), citing its impact on prognosis, treatment outcomes, and healthcare costs [31]. Despite these international standards, bleeding and other patient-centered measures were infrequently reported in the SRs/MAs analyzed in our study. Recommendations for SRs/MAs on DES for CAD To address this issue, we propose that the development of a comprehensive, widely accepted COS for CAD is essential. Incorporating COS guidance into the protocol registration stage may improve consistency in outcome reporting and enhance the reliability and comparability of future SRs/MAs. While technical endpoints are vital for evaluating device safety and performance, their application in real-world clinical decision-making is limited by the lack of integration with patient-relevant outcomes. Future research should adopt a more holistic approach, combining patient-reported outcomes with technical measures to develop an outcome framework that reflects clinical utility, device performance, and patient-centered care. In fact, a multisociety taskforce coordinated by the Heart Valve Society (e.g. the American Heart Association (AHA), the American College of Cardiology (ACC), the European Association for Cardio-Thoracic Surgery (EACTS), the European Society of Cardiology (ESC), etc.) have developed the first patient-related (International Consortium for Health Outcomes Measurement (ICHOM)) standard dataset for reporting outcomes in heart valve disease [32]. This dataset which focuses on outcome measurement regardless of treatment, moving from procedure- to patient-centered outcomes can facilitate global standardization of outcome measurement, allow meaningful comparison between health care systems and evaluation of clinical practice guidelines, and eventually improve patient care [32]. Strengths and limitations This study has several strengths. It is, to our knowledge, the first comprehensive analysis of outcome specification in SRs/MAs evaluating DES for CAD. The inclusion of 271 reviews spanning two decades offers a broad and representative overview of current practices. However, certain limitations should be acknowledged. First, in classifying outcomes according to the COMET framework, composite endpoints such as MACE, MACCE, TLF, DOCE, and POCE—which span multiple domains—were categorized primarily under adverse events due to the predominance of such components in their definitions. This may have influenced the apparent distribution of outcomes across domains. Second, the search was limited to English and Chinese databases. Consequently, grey literature or studies published in other languages, unpublished dissertations and conference abstracts were not included, which may introduce publication bias. Conclusions In summary, the inconsistent specification and limited standardization of primary outcomes in SRs/MAs of DES for CAD may undermine evidence quality. The pathway from “unspecified primary outcomes” to “lack of standardization” to an “incomplete core outcome set” suggests a need for methodological improvement in outcome reporting. Supplementary Information Supplementary Material 1. Supplementary Material 2. Supplementary Material 3. Supplementary Material 4. Acknowledgements This work was supported by the Hospital Pharmaceutical Research Project funded by the China Pharmaceutical Association (CPA-Z05-ZC-2023-002). We would like to thank the Drug Evaluation Center of Guizhou Province. Abbreviations Systematic Reviews Meta-Analyses Randomized Controlled Trials Non-Randomized Controlled Trials Cohort Studies Observational Studies Drug-Eluting Stents Bare-Metal Stents Drug-Eluting Balloons Drug-Coated Balloons Coronary Artery Bypass Grafting Cardiovascular Diseases Coronary Artery Disease Acute Coronary Syndrome Core Outcome Measures in Effectiveness Trials Consolidated Standards of Reporting Trials International Prospective Register of Systematic Reviews Grading of Recommendations Assessment, Development and Evaluation International Consortium for Health Outcomes Measurement Core Outcome Set Newcastle-Ottawa Scale Seattle Angina Questionnaire Patient Health Questionnaire Corrected Covered Area Disability-Adjusted Life Years Medical Subject Headings Kappa statistic Interquartile Range Odds Ratio Confidence Interval Major Adverse Cardiovascular Events Major Adverse Cardiovascular and Cerebrovascular Events Patient-Oriented Composite Endpoints Device-Oriented Composite Endpoints Target Vessel Revascularization Target Lesion Revascularization Target Vessel Failure Target Lesion Failure Myocardial Infarction Stent Thrombosis Authors’ contributions Xiaoqing Lin, Songsong Tan, Yuezi Huang, Jiaojiao Zhao, Jiankun Yao, Die Han, Renlan Luo, Huang Wu and Jiaxing Zhang collected the data. Jiaxing Zhang and Songsong Tan involved in statistical analysis. Xiaoqing Lin, Songsong Tan, and Jiaxing Zhang drafted the manuscript. Jiaxing Zhang, Junjie Lan, Rui He, and Wenyi Zheng revised the final manuscript. Rui Zhang, Huaye Zhao, Linfang Hu, Jiaxue Wang, and Shuimei Sun conceived and designed the study, performed critical revision of the manuscript for important intellectual content, and approved the final version of the manuscript to be published. Funding This work was supported by the Hospital Pharmaceutical Research Project funded by the China Pharmaceutical Association (CPA-Z05-ZC-2023-002). But the funding did not involve in the study design, the collection, analysis and interpretation of data, writing of the report, or the decision to submit the article for publication. Data availability All data generated or analyzed during this study are included in this published article and its supplementary information files. Declarations Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Xiaoqing Lin and Songsong Tan are co-first authors and contributed equally to this work. References Associated Data Supplementary Materials Supplementary Material 1. Supplementary Material 2. Supplementary Material 3. Supplementary Material 4. Data Availability Statement All data generated or analyzed during this study are included in this published article and its supplementary information files.