Background: Coronavirus disease 2019 (COVID-19) has spread worldwide rapidly. However, the effects of asthma, asthma medication and asthma severity on the clinical outcomes of COVID-19 have not yet been established. Methods: The study included 7590 de-identified patients, who were confirmed to have COVID-19 using the severe acute respiratory syndrome coronavirus 2 RNA-PCR tests conducted up to May 15, 2020; we used the linked-medical claims data provided by the Health Insurance Review and Assessment Service. Asthma and asthma severity (steps suggested by the Global Initiative for Asthma) were defined using the diagnostic code and history of asthma medication usage. Results: Among 7590 COVID-19 patients, 218 (2.9%) had underlying asthma. The total medical cost associated with COVID-19 patients with underlying asthma was significantly higher than that of other patients. Mortality rate for COVID-19 patients with underlying asthma (7.8%) was significantly higher than that of other patients (2.8%; p<0.001). However, asthma was not an independent risk factor for the clinical outcomes of COVID-19 after adjustment, nor did asthma medication use and asthma severity affect the clinical outcomes of COVID-19. However, use of oral short-acting β ₂ -agonists was an independent factor to increase the total medical cost burden. Patients with step 5 asthma showed significant prolonged duration of admission compared to those with step 1 asthma in both univariate and multivariate analysis. Conclusions: Asthma led to poor outcomes of COVID-19; however, underlying asthma, use of asthma medication and asthma severity were not independent factors for poor clinical outcomes of COVID-19, generally.

Purpose: The forced mid-expiratory flow (FEF_25-75%) value is a potentially sensitive marker of obstructive peripheral airflow. We aimed to assess whether FEF_25-75% can be an early predictor of chronic obstructive pulmonary disease (COPD). Patients and Methods: Between July 1, 2007 and June 31, 2009, we identified 3624 patients who underwent a pulmonary function test (PFT) in Gangnam Severance Hospital. We selected 307 patients aged over 40 years without COPD who had normal PFT results at baseline and who had follow-up PFT records more than 1 year later. A FEF_25-75% z-score less than − 0.8435 was considered low. We defined COPD as a forced expiratory volume in one second/forced vital capacity value of less than 0.7 before July 31, 2019. Results: Among 307 patients, 91 (29.6%) had low FEF_25-75% at baseline. After 10 years, the incidence rate of COPD in the low FEF_25-75% group was significantly higher than that in the normal FEF_25-75% group (41.8% vs 7.4%; P-value< 0.001). The Cox proportional hazard model showed that age (hazard ratio [HR] 1.09; P-value< 0.001), smoking status (occasional smoker HR, 4.59; P-value< 0.001 and long-term smoker HR, 2.18; P-value=0.023), and low FEF_25-75% (HR, 3.31; P-value< 0.001) were predictive factors for the development of COPD. Conclusion: The FEF_25-75% value in patients with normal lung function is a useful predictor for the development of COPD. We should carefully monitor patients who present with low FEF_25-75% values, even if they have normal lung function.

Improving predictive models for intensive care unit (ICU) inpatients requires a new strategy that periodically includes the latest clinical data and can be updated to reflect local characteristics. We extracted data from all adult patients admitted to the ICUs of two university hospitals with different characteristics from 2006 to 2020, and a total of 85,146 patients were included in this study. Machine learning algorithms were trained to predict in-hospital mortality. The predictive performance of conventional scoring models and machine learning algorithms was assessed by the area under the receiver operating characteristic curve (AUROC). The conventional scoring models had various predictive powers, with the SAPS III (AUROC 0.773 [0.766–0.779] for hospital S) and APACHE III (AUROC 0.803 [0.795–0.810] for hospital G) showing the highest AUROC among them. The best performing machine learning models achieved an AUROC of 0.977 (0.973–0.980) in hospital S and 0.955 (0.950–0.961) in hospital G. The use of ML models in conjunction with conventional scoring systems can provide more useful information for predicting the prognosis of critically ill patients. In this study, we suggest that the predictive model can be made more robust by training with the individual data of each hospital.

Underlying diseases might be risk factors for poor prognosis in patients with coronavirus disease (COVID-19); however, we still do not know whether these diseases are independent factors affecting prognosis, which type of underlying diseases are risk factors, and which type of clinical outcomes are affected. We retrospectively reviewed cohort data from 7,590 de-identified patients with COVID-19 who were diagnosed using severe acute respiratory syndrome-coronavirus-2 RNA polymerase chain reaction test up to May 15, 2020. We used linked-medical claims data provided by the Health Insurance Review and Assessment Service in South Korea. Underlying diseases were identified using the diagnostic codes in the patients’ files from January 1, 2019 to December 31, 2019. The total mortality rate was 3.0% in patients with COVID-19. After adjusting for age, sex, and concomitant chronic conditions, we found that congestive heart failure, chronic pulmonary diseases, diabetes without chronic complications, renal diseases, and malignancy were factors that significantly increased the cost of treatment. Cerebrovascular disease, chronic pulmonary disease, and paralysis were found to be independent factors significant in prolonging hospital stay. Diabetes with chronic complications was independently associated with intensive care unit admission. In addition, underlying congestive heart failure (odds ratio [OR], 1.724; P = 0.003), dementia (OR, 1.598; P = 0.012), diabetes with and without chronic complications (OR, 1.821; P = 0.002 and OR, 1.518; P = 0.022, respectively), renal disease (OR, 2.299; P = 0.002), and malignancy (OR, 1.529; P = 0.039) were significant factors associated with death, even after adjustments. Underlying diseases were significant independent factors of the poor prognosis in patients with COVID-19. The effects were variable according to the type of underlying disease and clinical outcome. Therefore, patients with COVID-19 with underlying diseases should be monitored more closely because they are more at risk of a poor prognosis.

• Tregs lead immune-evasive TME of early lung cancer of never smoker. • Depletion of γδT and NK cells and infiltration of B cells begins in early TME. • Early lung cancer cells were characterized by dysregulated surfactant pathway. • CAFs show enrichment of gene sets that inhibit vascular formation. In tumor tissue, tip-like endothelial cells begin to be replaced with immature ones.

Background: Hyperprogressive disease (HPD) is a paradoxical acceleration of tumor growth after immune checkpoint inhibitor (ICI) treatment. This study aimed to identify the risk factors and to present a predictive model for HPD in patients treated with ICIs.Methods: A total of 78 non‐small cell lung cancer (NSCLC) cases, treated with at least two cycles of ICIs who underwent computed tomography (CT) for response assessment were recruited into the study from January 2016 to August 2019. HPD was defined by the following criteria: (i) time‐to‐treatment failure <2 months; (ii) a 50% increase in the sum of target lesion diameters; (iii) new development of at least two lesions in an already involved organ; (iv) appearance of a new organ lesion; and (v) a decrease in ECOG PS 2.Results: Of the 78 total patients, 15 (19.2%) had HPD. The risk factors of HPD were age; primary lesion size; and metastases in the contralateral lung, pleura, liver, and bone in multivariable logistic regression (odds ratio [OR]; 0.9038, 1.6619, 28.5913, 23.8264, 14.5711, and 20.1533, respectively, all P‐values < 0.05). By using these risk factors, we developed a prediction model for HPD and the area under the receiver operating characteristic curve of the model was 0.9556 (95% confidence interval [CI]: 0.9133–0.9978).Conclusions: HPD is relatively common and associated with a grave clinical outcome, requiring a careful monitoring in lung cancer patients treated with ICIs. Moreover, risk factors such as age, size of tumor and number of various metastatic lesions should be taken into consideration before ICI administration.Key points: Significant findings of the study: Age, primary lesion size, and number of metastases are risk factors of HPD. HPD is strongly associated with poor prognosis. HPD during ICI use needs comprehensive monitoring.What this study adds: This is the first study to develop a prediction model. The area under the curve of the prediction model for HPD was 0.9556.