Background:Lung age (LA) that expresses lung function as a form of biological age and facilitates clinical interpretation. Bronchodilator responsiveness (BDR) is important for assessing airway reversibility, but its influence on LA remains unclear. This study aimed to evaluate the impact of BDR on LA and to explore its clinical implications. Methods:In this cross-sectional study, we analyzed participants with respiratory disease and/or symptoms who underwent spirometry and BDR testing between January 2023 and December 2024. LA was calculated using a reference equation established from healthy, non-smoking Chinese adults. Participants were categorized into three groups: normal [the difference between lung age and age (DeltaLA) ≤0], normally high [DeltaLA >0 and ≤ upper limit of normal for DeltaLA (DeltaLAULN)], and abnormally high (DeltaLA > DeltaLAULN). A random forest model was used to identify predictors of post-LA improvement. Nonlinear regression was employed to evaluate associations between DeltaLA and spirometric parameters. Receiver operating characteristic (ROC) analysis was performed to evaluate the ability of difference between ULN and LA (ULNLA) and spirometric measurements to predict a post-bronchodilator forced expiratory volume in one second to forced vital capacity (FEV1/FVC) ratio <0.7. Results:Nine thousand three hundred and sixteen participants (mean age 58.6±12.9 years; 69.7% male) were included. After bronchodilator, LA significantly improved and 34.03% of those in normally high LA improved to normal group (all P<0.001); 21.83% of participants in the abnormally high LA group had a significant BDR, compared with 3.68% and 4.02% in the normal and normally high LA groups, respectively (P<0.001). DeltaLA was correlated with FEV1, forced expiratory flow at 50% of FVC (FEF50%), forced expiratory flow at 75% of FVC (FEF75%), and FVC (r=-0.80 to -0.46, all P<0.001). The ability of ULNLA to predict a post-bronchodilator FEV1/FVC ratio <0.7 was comparable to FEV1 (area under the curve: 0.84 vs. 0.85). Conclusions:Pre-bronchodilator LA alone may result in misclassification, whereas post-bronchodilator LA provides a more accurate evaluation. ULNLA demonstrates predictive ability of airflow limitation is comparable to FEV1% predicted.

Invasive pulmonary aspergillosis (IPA) is a life-threatening infection with high mortality, often misdiagnosed due to non-specific symptoms. The lack of effective early prediction models delays treatment and worsens outcomes. To identify independent IPA risk factors and develop a clinically actionable nomogram for early prediction, we conducted a nested case-control study within a 10-year cohort of 27 100 pulmonary infection patients. The cohort was split into training (70%) and testing (30%) sets. In the training set, 1002 IPA cases (proven or probable, as defined by European Organization for Research and Treatment of Cancer and Mycosis Research Group Education and Research Consortium 2020 criteria) were included, alongside 2004 randomly selected pneumonia controls (1:2 ratio). A nomogram was developed using multivariable logistic regression in the training set and was evaluated by held-out internal validation in the testing set. Model performance was primarily assessed via area under the curve (AUC), calibration, and decision curve analysis (DCA). Nine independent IPA predictors were identified: bronchiectasis (odds ratio [OR]= 7.07), pulmonary tuberculosis (OR = 2.20), diabetes (OR = 2.09), positive serum galactomannan (GM) test (OR = 1.76), mechanical ventilation (OR = 1.74), connective tissue disease (OR = 1.73), positive serum (1,3)-β-d-glucan (G) test (OR = 1.48), sputum (OR = 1.33), and neutrophil-to-lymphocyte ratio (OR = 1.01). The nomogram demonstrated moderate stable discrimination (training AUC: 0.73; testing AUC: 0.75), with excellent calibration (Brier scores: 0.183 and 0.110). DCA confirmed clinical utility across wide risk thresholds. In conclusion, this novel nine-variable nomogram using routine clinical data provides a practical tool for early IPA risk estimation, potentially guiding timely decisions and improving outcomes. External validation is warranted.

Impulse oscillometry system (IOS) is an effort-independent pulmonary function testing technique that assesses respiratory system mechanics during tidal breathing and has gained increasing interest as a complementary modality to traditional pulmonary function tests (PFTs), while its physiological relevance and clinical role alongside conventional tests remain incompletely defined. This narrative literature review synthesises evidence on the correlations between IOS parameters and traditional PFT indices, summarises clinical applications across major chronic respiratory diseases, and discusses emerging developments including artificial intelligence-based approaches; the review is based on studies published since 2010 that compare IOS with spirometry, body plethysmography, and diffusing capacity of the lung for carbon monoxide (DLCO) in chronic obstructive pulmonary disease (COPD), asthma, and interstitial lung disease (ILD). Correlation studies indicate that IOS resistance-related parameters are more closely associated with obstructive ventilatory impairment, particularly in COPD and asthma, whereas reactance-related parameters show stronger associations with restrictive ventilatory abnormalities, lung volumes, and diffusion impairment, especially in ILD. Across disease entities, R5-20 and resonant frequency consistently demonstrate higher sensitivity for detecting small airway dysfunction. Clinically, IOS provides value in assessing bronchodilator responsiveness, evaluating asthma control, detecting small airway involvement when spirometry is preserved, and monitoring exposure-related airway effects. Artificial intelligence-based models integrating multidimensional IOS data further highlight its potential in disease screening and classification. Overall, IOS provides complementary physiological and clinical information beyond traditional PFTs, and further large-scale, multicentre studies and methodological standardisation are needed to support its broader clinical implementation.

INTRODUCTION:The dysregulation of both reduced (IL-33red) and oxidised (IL-33ox) interleukin (IL)-33 has been implicated in chronic obstructive pulmonary disease (COPD) inflammation and remodelling processes. Tozorakimab, an anti-IL-33 monoclonal antibody, inhibits both IL-33red and IL-33ox activity. METHODS AND ANALYSIS:The tozorakimab LUNA programme comprises four ongoing, multicentre, randomised, double-blind, parallel-group, placebo-controlled, phase III studies evaluating the efficacy and safety of tozorakimab in participants with symptomatic COPD and a history of exacerbations receiving optimised inhaled therapy. In OBERON (NCT05166889) and TITANIA (NCT05158387), 1132 and 1172 participants, respectively, were randomised to receive tozorakimab 300 mg every 4 or 8 weeks or placebo for 52 weeks. In MIRANDA (NCT06040086), 1454 participants were randomised 3:2 to tozorakimab 300 mg every 2 weeks or placebo for ≥52 weeks. In PROSPERO (NCT05742802), participants who completed treatment with tozorakimab in OBERON or TITANIA will continue treatment for an additional 28 weeks or 52 weeks; participants recruited from the placebo arm will be re-randomised 1:1 to tozorakimab or placebo. The primary endpoint in OBERON, TITANIA and MIRANDA is the annualised rate of moderate-to-severe COPD exacerbations; the primary endpoint in PROSPERO is the annualised rate of severe COPD exacerbations. For all studies, the primary endpoint will be first assessed in former smokers, then in current and former smokers. Secondary key endpoints include measures of lung function, respiratory symptoms, health status and safety. The tozorakimab LUNA programme is assessing the efficacy and safety of tozorakimab in participants with COPD. It is the largest pivotal programme of any biologic therapy in COPD to date. ETHICS AND DISSEMINATION:The protocols were approved by independent ethics committees and institutional review boards. Results of the studies will be submitted to the EU Clinical Trials Information System within a year from the global end of trial data in all participating countries and will be published or presented at scientific meetings. TRIAL REGISTRATION NUMBER:OBERON: NCT05166889; TITANIA: NCT05158387; MIRANDA: NCT06040086; PROSPERO: NCT05742802.

BACKGROUND:The chronic obstructive pulmonary disease (COPD) guidelines recommend selecting initial treatment based on the severity of respiratory symptoms or exacerbation history. However, some patients still experience clinically important deterioration (CID) despite this strategy. Therefore, we aimed to identify the risk factors to guide initial treatment, focusing on mild-to-moderate COPD patients whose annualized CID cannot be controlled by a single-bronchodilator. METHODS:Patients in the tiotropium group from a randomized controlled trial were included. Over 2 years, annualized CID was defined as forced expiratory volume in one second (FEV1) decline at least 100 ml, COPD assessment test (CAT) increasing at least 2 points, or having experienced at least two moderate or one severe acute exacerbations of COPD (AECOPD) in this study. Random coefficients model within a bayesian framework was adapted to estimate the change in FEV1 and CAT scores. The logistic regression analysis was adapted to detect the risk factors. The risk of annualized CID was evaluated using the incidence risk ratio (IRR) for patients with risk factors. RESULTS:Of 312 patients with mild-to-moderate COPD receiving tiotropium, 29.2% still experienced annualized CID. The proportions of patients who developed annualized CID regarding FEV1, CAT, and AECOPD was 20.5, 8.0, and 6.1%, respectively. Post-bronchodilator FEV1:FVC ≤ 60% (OR = 1.73, 95%CI: 1.06-2.84) and smoking pack-years ≥50 (OR = 1.95, 95%CI: 1.11-3.41) were associated with higher risk of experiencing annualized CID. The IRR of annualized CID was increased with more risk factors at baseline. CONCLUSION:Patients with more prominent airflow limitation and higher smoking pack-years warrant intensified initial treatment to control their annualized CID.

OBJECTIVE:Quality control (QC) of spirometry is a critical yet challenging prerequisite for accurate pulmonary function tests (PFTs), especially in primary care settings where skilled personnel are often unavailable. This study aims to develop an intelligent evaluation model to automatically identify anomaly types in spirometry reports, ensuring the reliability of diagnostic data.&#xD;&#xD;Approach: We proposed a hybrid model, which integrates an artificial neural network (ANN) with wavelet analysis and integration of Convolutional Neural Network and Long Short-Term Memory models (CNN-LSTM). The model was trained and validated using over 3,000 curves collected from the First Affiliated Hospital of Guangzhou Medical University.&#xD;&#xD;Main results: The proposed model achieved an average accuracy of 96.73\% on the validation set and 95.84\% agreement on the test set. Comparative analysis shows that our method outperforms existing benchmarks in Precision (0.965), Recall (0.971), and F1 score (0.968). The model successfully identifies a wide range of anomalies, including cough, glottic closure, and initial leakage, with high consistency.&#xD;&#xD;Significance: This approach provides a high-precision, near real-time diagnostic tool for spirometry quality assessment. By simplifying data management and standardizing evaluation processes, the model supports primary healthcare institutions in enhancing testing quality and diagnostic reliability, particularly in resource-limited environments.

Background:History of exacerbations is a predictor of future exacerbations in COPD but there are also predictors that are independent of exacerbation history. However, it is unclear whether their relative contribution is consistent across different degrees of airflow limitation. Methods:This analysis used data from COMPASS, a prospective study in COPD. Baseline demographics, clinical history, spirometry and patient-reported outcomes were collected. Multivariable models were created to predict moderate or severe exacerbations in the 18 months after baseline. Covariates included forced expiratory volume in 1 s (FEV1) % predicted, Global Initiative for Chronic Obstructive Lung Disease (GOLD) grade, modified Medical Research Council (mMRC) and COPD Assessment Test (CAT) scores, and exacerbation history. Goodness of fit was tested using C-statistics. Results:At baseline there were 1696 patients; 89.6% males, 46.9% current smokers, mean±sd age of 65.4±7.5 years, post-bronchodilator FEV1 66.6±20.1% predicted and 0.5±1.0 moderate/severe exacerbations in the prior year. Over 18 months, 17.8% of patients had ≥1 moderate/severe exacerbation. The best fit model identified six independent variables, C-statistic 0.739. Subgroup analysis into GOLD grades I, II and III+IV combined showed different predictor patterns. In grade I, history of moderate exacerbations was the strongest predictor, together with chronic bronchitis and gastro-oesophageal reflux. In grades III+IV, only history of severe exacerbations and mMRC score were significant. Grade II showed an intermediate picture in which severe exacerbations, chronic bronchitis and gastro-oesophageal reflux were all significant. Conclusions:There are multiple predictors of COPD exacerbations, which differ between GOLD grades. Future predictive models for exacerbation risk should take this into account.

BACKGROUND:Many patients with chronic obstructive pulmonary disease (COPD) have exacerbations despite receiving standard-of-care inhaled maintenance therapy. Dysregulated interleukin-33 signaling is implicated in the pathogenesis of COPD. Tozorakimab is a monoclonal antibody that inhibits the activity of interleukin-33. METHODS:In two replicate phase 3 trials (OBERON and TITANIA), we enrolled adults with COPD who were current or former smokers and had a history of exacerbations in the previous year despite receiving stable standard-of-care inhaled maintenance therapy. There were no eligibility criteria related to blood eosinophil count. Patients were randomly assigned to receive add-on subcutaneous tozorakimab (300 mg) or placebo every 4 weeks for 52 weeks. The primary end point was the annualized rate of moderate or severe exacerbations that occurred over a 52-week period among former smokers, and the first key secondary end point was the annualized rate of moderate or severe exacerbations in the overall population. Safety was also assessed. RESULTS:The overall population in OBERON included 446 patients in the tozorakimab group and 431 in the placebo group, and in TITANIA included 438 in the tozorakimab group and 435 in the placebo group. The annualized rate of moderate or severe exacerbations among former smokers was 1.34 events in the tozorakimab group and 1.90 events in the placebo group (rate ratio, 0.71; 95% confidence interval [CI], 0.57 to 0.88; P = 0.002) in OBERON, and 1.37 events and 2.07 events, respectively (rate ratio, 0.66; 95% CI, 0.55 to 0.80; P<0.001), in TITANIA. In the overall population, the annualized rate of moderate or severe exacerbations was 1.41 events in the tozorakimab group and 2.00 events in the placebo group (rate ratio, 0.70; 95% CI, 0.58 to 0.85; P<0.001) in OBERON, and 1.44 events and 2.03 events, respectively (rate ratio, 0.71; 95% CI, 0.59 to 0.84; P<0.001), in TITANIA. Adverse events occurred in 70.4% of the patients in the tozorakimab group and in 77.2% of those in the placebo group in OBERON, and in 80.1% and 79.8%, respectively, in TITANIA. CONCLUSIONS:Among patients with COPD, treatment with tozorakimab resulted in a significantly lower rate of moderate or severe exacerbations than placebo in the cohort of former smokers and in the overall population of current and former smokers. (Funded by AstraZeneca; OBERON ClinicalTrials.gov number, NCT05166889; TITANIA ClinicalTrials.gov number, NCT05158387.).

Spirometry plays a key role in diagnosing respiratory diseases, but its accuracy often falls short of clinical expectations. While deep learning models have shown promise in automating spirometry analysis, challenges persist. Spirometry curves are vulnerable to noise from minor signals, which can lead to diagnostic errors. Moreover, the high computational demands of current algorithms limit their use in clinical practice. To address these issues, we present a novel approach that integrates n-order adaptive Fourier decomposition (AFD) with deep learning techniques to enhance quality control in spirometry analysis. AFD improves the resolution and processing of flow-volume (F-V) curves, effectively minimizing noise. By leveraging the strengths of deep learning models alongside AFD, our method accurately detects small and complex abnormalities in spirometry data. Ablation studies show that our method outperforms traditional approaches, raising the mean average precision (mAP) from 89.5% to 95.5%. Furthermore, the model's lightweight design, achieved through computational optimizations and structural simplifications, enables efficient deployment in various clinical settings, improving diagnostic accuracy and accessibility.

Background Lung age is an easily understandable index developed to quantify the physiological function level of the lungs, but its role in identifying and managing high-risk populations and mild COPD remined unclear. Objective: We aimed to observe the baseline characteristics and progression of lung age in healthy individuals, chronic bronchitis (CB), and COPD patients, and to assess the association between lung age and annualized Clinically Important Deterioration (CID). Methods: The COMPASS cohort is a multi-centers prospective study observating COPD patients for 2.5 years in China, data was collected at planned annual visits. The lung age is calculated using a formula that incorporates the measured values of FEV1, Forced Expiratory Flow 50% (FEF50), and FEF75. A lung age greater than the actual age indicates a deterioration in lung function.1 Mixed linear models and ANOVA were employed to assess the annualized change in lung age. Logistic regression was also used to assess the association between annualized changes in lung age and the occurrence of annualized CID (i.e., decrease exceeding 100ml/year in Forced Expiratory Volume in 1 Second [FEV1], increase greater than 2 points/year in COPD Assessment Test [CAT], or 2 moderate to severe Acute Exacerbation of COPD [AECOPD]/year). Results: A total of 1,488 participants were enrolled, including 104 healthy individuals, 141 chronic bronchitis patients, and 1,243 COPD patients. Lung age at baseline was much less in healthy individuals (56.4), CB (61.0) and mild COPD (66.2) than in moderate to very severe COPD (e.g., GOLD3-4 119.9). The annual increase in lung age was slowly (i.e.,1.5 years/year and 1.4 years/year) for healthy individuals and moderate-to-severe COPD patients, respectively. In contrast, CB and mild COPD patients exhibited larger annual increase for 2.9 years/year and 2.6 years/year, respectively (P for ANOVA < 0.001). During follow-up, the percentage of patients CID positive by FEV1, CAT, and exacerbations were 9.1%, 17.3% and 26.7%, respectively. For each unit increase in lung age per year, the risk of experiencing CID in chronic bronchitis, GOLD stage 1, 2, and 3/4 increased by 329%, 163%, 22%, and 107%, respectively. Conclusion: Lung age deteriorates rapidly in CB and mild COPD patients and is associated with greater risk of CID. It can serve as a tool for identifying patients at high risk of deterioration. Figure 1. The association of increase in lung age and clinical important deterioration (CID) in different group.

Background:COPD and cardiovascular disease (CVD) are leading causes of death with overlapping and syndemic pathophysiological interactions. Inhaled triple therapies containing inhaled corticosteroids (ICS), long-acting muscarinic antagonists (LAMA) and long-acting β2-agonists (LABA) reduce COPD exacerbation rates and improve lung function versus dual LAMA/LABA therapy. The effect of inhaled triple therapies on combined cardiac and pulmonary (i.e., cardiopulmonary) events in people with COPD and elevated cardiopulmonary risk has not been prospectively tested in randomised clinical trials. Methods:THARROS is a multinational, event-driven cardiopulmonary outcomes trial evaluating budesonide/glycopyrronium/formoterol fumarate dihydrate (BGF) triple therapy versus glycopyrronium/formoterol fumarate dihydrate dual therapy in patients with COPD and elevated cardiopulmonary risk not using ICS-containing maintenance therapy. Eligibility requirements include symptomatic COPD (COPD Assessment Test scores ≥10) without a requirement for prior COPD exacerbations, blood eosinophils ≥100 cells·mm-3, established CVD or CVD risk based on clinical characteristics, clinical risk scores or imaging-based risk criteria. The composite primary end-point is time to first severe cardiac or COPD event and includes three event types, including severe cardiac events (heart failure acute healthcare visit/hospitalisation, myocardial infarction hospitalisation), severe COPD exacerbations and cardiopulmonary death. Approximately 5000 patients will be randomised to achieve 632 participants with ≥1 primary severe adjudicated cardiopulmonary event. Conclusion:This first-of-its-kind cardiopulmonary outcomes trial will determine the effect of BGF on a novel composite end-point comprising severe cardiopulmonary events in a broad COPD population with elevated cardiopulmonary risk not currently using ICS-containing maintenance therapy.

RATIONALE: There is evidence of type 2 inflammation in a subset of patients with COPD. Dupilumab, a fully human monoclonal antibody, blocks the shared receptor component for interleukin (IL)-4 and IL-13, key and central drivers of type 2 inflammation. In BOREAS (NCT03930732) and NOTUS (NCT04456673), patients with COPD, moderate-to-severe airflow limitation, and type 2 inflammation (≥300 cells/µL blood eosinophils at screening) on background triple therapy, who received dupilumab 300 mg every 2 weeks or placebo for 52 weeks, had a statistically significant reduction in exacerbations and improved lung function. Safety was consistent with the known dupilumab safety profile. This analysis examined multiple lung function parameters in the pooled population and current and former smokers from both trials. METHODS: Change in pre- and post-bronchodilator forced expiratory volume in 1 second (FEV1), post-bronchodilator FEV1/forced vital capacity (FVC) ratio, and post-bronchodilator forced expiratory flow 25-75% of vital capacity (FEF25-75%) over time were analyzed. RESULTS: In the intention-to-treat (ITT) population, patients received dupilumab (n = 938) or placebo (n = 936); 830 patients from each group had the opportunity to reach Week 52. The ITT population was used for subgroup (current and former smokers) analysis. In the pooled ITT population who had the opportunity to reach Week 52, the baseline mean (standard deviation) pre-bronchodilator FEV1 was 1.32 (0.47) and 1.35 (0.48) L; post-bronchodilator FEV1 was 1.41 (0.47) and 1.44 (0.49) L; post-bronchodilator FEV1/FVC ratio was 0.49 (0.12) and 0.49 (0.12); and post-bronchodilator FEF25-75% was 0.56 (0.33) and 0.56 (0.33) L/s for dupilumab and placebo, respectively. By Week 12, dupilumab vs placebo improved pre-bronchodilator FEV1 (least squares [LS] mean difference [95% CI] 83 [51, 114] mL), post-bronchodilator FEV1 (LS mean difference [95% CI] 72 [40, 105] mL), post-bronchodilator FEV1/FVC (LS mean difference [95% CI] 16 [9, 23]), and post-bronchodilator FEF25-75% (LS mean difference [95% CI] 80 [44, 116] mL/s). Improvements were sustained up to Week 52. In the ITT population, the LS mean difference (95% CI) vs placebo in pre-bronchodilator FEV1 at Week 12 was 89 (52, 127) in former smokers (n = 661 dupilumab; n = 654 placebo)and 64 (20, 109) mL in current smokers (n = 275 dupilumab; n = 282 placebo). CONCLUSIONS: In 2 pivotal phase 3 trials in patients with COPD and type 2 inflammation, dupilumab improved lung function by Week 12 and sustained the improvement throughout the study period.

With the development and market launch of several new domestic lung diffusing capacity testing instruments, the clinical reliability of the MeHow MeAir 9000 spirometer, featuring a novel turning valve structure, needs to be validated. To evaluate the clinical reliability of lung diffusing capacity measurements using the MeHow MeAir 9000 spirometer. This study included 166 participants: 30 healthy individuals, 68 with interstitial lung disease (ILD), and 68 with chronic obstructive pulmonary disease (COPD). Using a crossover design, participants underwent lung diffusing capacity tests with both the MeHow MeAir 9000 and Jaeger MasterScreen Diffusion spirometers, following the 2017 ERS/ATS standards. The primary indicator was the diffusing capacity for carbon monoxide (DLCO), with the diffusion capacity of carbon monoxide as a percentage of the predicted value (DLCO

Background Airway mucus hypersecretion is a key pathophysiological feature in many respiratory diseases and could lead to airway obstruction and repeated infections, consequently accelerating disease progression, which impacts on pulmonary function and quality of life (QoL), highlighting the importance of mucolytic therapy targeting airway mucus hypersecretion.Objectives To investigate the safety and efficacy of inhalable ambroxol hydrochloride aerosol for adult patients with respiratory diseases.Design An open-label, single-arm, multicentre postmarketing surveillance study.Methods Adult patients with acute or chronic respiratory diseases were eligible to receive aerosol inhalation of ambroxol hydrochloride (3 mL and 7.5 mg/mL) using a nebuliser two times per day given at least 6 hours apart between doses. The treatment lasted for a maximum of 7 days. The primary safety outcome was the frequency and severity of adverse events (AEs), and the primary efficacy outcome was changes in sputum scale scores.Results Among 1201 eligible patients, 1192 received study medication and were included in the full analysis set and the safety set. Any grade AEs occurred in 16.3% of the patients, including serious AEs in four (0.3%) patients. The three most frequent AEs were respiratory symptoms and signs (1.5%), nausea and vomiting (0.8%) and digestive tract symptoms and signs (0.7%). In the full analysis set, the patients showed a mean reduction of 77.6% (95% CI, 75.9% to 79.3%) in the sputum scale score at the end of treatment, with a mean difference of −1.7±0.7 from baseline (p<0.001).Conclusion Inhalable ambroxol hydrochloride aerosol is well tolerated and effective in easing expectoration and alleviating cough, reducing sputum and improving the QoL of adult patients with acute and chronic respiratory diseases.Trial registration number ChiCTR2100043736.

Background:The rate of diagnosis for chronic obstructive pulmonary disease (COPD) is low worldwide. Quantitative computed tomography (QCT) parameters add value to quantify alterations in airway and lung parenchyma for COPD. This study aimed to assess the performance of QCT features in COPD detection using a whole-lung inspiratory CT model. Methods:This multicenter retrospective study was performed on 4106 participants. The derivation cohort containing 1950 participants who enrolled in Guangzhou communities from August 2017 to December 2019, was separated for training and internal validation cohorts, and three external validation cohorts containing 1703 participants were recruited from the public hospitals (Cohort 1: the First Affiliated Hospital of Guangzhou Medical University; Cohort 2: Xiangyang central hospital; Cohort 3: the Second Affiliated Hospital of Xi'an Jiaotong University) in China between April 2017 and May 2024. Questionnaire information, CT reports, and QCT features derived from inspiratory CT were extracted for model development. A novel multimodal framework using eXtreme gradient boosting and hybrid feature selection was established for COPD detection. National Lung Screening Trial (NLST) cohort (n = 453) was applied to validate the multiracial extrapolation and robustness on low-dose CT scans. Findings:The QCT model (referred to as AutoCOPD) with ten features achieved the highest AUC of 0·860 (95% CI: 0·823-0·898) in the internal validation cohort, and showed excellent discrimination when externally validated [Cohort 1: AUC = 0·915 (95% CI: 0·898-0·931); Cohort 2: AUC = 0·903 (95% CI: 0·864-0·943); Cohort 3: AUC = 0·914 (95% CI: 0·882-0·947); NLST: AUC = 0·881 (95% CI: 0·846-0·915)]. Decision curve analysis demonstrated that AutoCOPD was valuable across a range of COPD risk thresholds between 0·12 and 0·66 compared with intervention in all patients with COPD or no intervention. Interpretation:Heterogeneous COPD can be well identified using AutoCOPD (https://lwj-lab.shinyapps.io/autocopd/) constructed by a subset of only ten QCT features. It may be generalizable across clinical settings and serve as a feasible tool for early detecting patients with mild or asymptomatic COPD to reduce delayed diagnosis in routine practice. Funding:The National Natural Science Foundation of China, Guangzhou Laboratory, Natural Science Foundation of Guangdong Province, Guangzhou Municipal Science and Technology grant, State Key Laboratory of Respiratory Disease.

Rationale: Chronic obstructive pulmonary disease (COPD) has multiple impacts, so there is interest in using composite measures to assess treatment effects. Disease stability is emerging as a treatment target in COPD, and its inverse, the clinically important deterioration (CID), has been previously validated (Naya et al. Respir Res. 2018;19:222) and used as an outcome in clinical trials (Naya et al. ERJ Open Res. 2018;4:00047-2018; Maltais et al. Respir Res. 2019;20:238). CID has three components: 100 mL loss of forced expiratory volume in 1 second (FEV1), a minimum clinically significant deterioration (MCID) in a patient-reported outcome, and the occurrence of a moderate or severe exacerbation. Since clinical outcomes may depend on baseline disease severity, this analysis examined the frequency of CID in patients across a wide range of airflow limitations. Methods: Data are from COMPASS, a prospective 2.5-year cohort study of COPD in China (Liang et al. ERJ Open Res. 2021;7:00201). Measurements from baseline to 6 months were used to allow comparability with other studies. Changes in post-bronchodilator FEV1 and COPD Assessment Test (CAT) scores were measured. As in previous studies, the 2-unit MCID deterioration threshold for the CAT was used as its threshold for deterioration. The occurrence of a moderate or severe exacerbation was captured. Results: At 6 months, changes from baseline were available in 1649 participants across Global Initiative for Chronic Obstructive Lung Disease (GOLD) grades I-IV (Figure). The mean post-bronchodilator FEV1 at baseline was 66% predicted. The overall CID rate was very consistent across GOLD grades, with a range of 46-55% (Figure). However, whilst fewer patients experienced a 100 mL loss of FEV1 in GOLD III and IV compared to grades I and II, a greater number of GOLD III and IV patients had an exacerbation. The proportion of patients who deteriorated on the CAT was consistent across GOLD grades. Conclusions: There was a high level of consistency in CID rate in patients across a wide range of airflow limitation. The relative contribution of the CID components to the overall rate differed according to the degree of airflow limitation, although the percentage of patients with worsened CAT score was independent of baseline airflow limitation. These data support the use of CID in patients with mild to very severe COPD.