PSS-ILD lung-free transplantation survival prediction model and application

Abstract

The invention discloses a pSS-ILD lung-free transplantation survival prediction model and application, and relates to the technical field of pSS-ILD lung-free transplantation survival prediction. A training queue and an external verification queue are established, and pSS-ILD is evaluated through multidisciplinary discussion; the patients are divided into an F-pSS-ILD group and an NF-pSS-ILD group according to an HRCT mode, and the degree of imaging features, especially fibrosis scores, is quantified; analyzing and evaluating the relationship between clinical features and pSS-ILD prognosis by adopting single-factor Cox regression, and formulating a column diagram prediction model for predicting lung-free transplantation survival of pSS-ILD patients in 1, 3 and 5 years; the test of the column diagram prediction model adopts a subject working characteristic ROC curve, an area under the ROC curve and a consistency index to evaluate the prediction accuracy of the model. According to the method, a risk prediction model based on the four factors of age, oxygenation index, CA125 and fibrosis score is constructed, the model is a visual Nomogram visual model for evaluating the lung-free transplantation survival rate of 1, 3 and 5 years, a visual column graph model is provided, and a simple, visual and practical prediction tool is provided for clinicians.

Description

Technical Field

[0001] The present invention relates to the technical field of pSS-ILD lung transplant-free survival prediction, and in particular to a pSS-ILD lung transplant-free survival prediction model and application thereof. Background Art

[0002] Primary Sjögren's syndrome pSS is a chronic autoimmune disease characterized by lymphocyte infiltration of exocrine glands, which can lead to symptoms such as dry eyes (dry eyes) and dry mouth (xerostomia). However, the systemic nature of pSS often leads to a range of extraglandular manifestations, the most important of which is lung involvement. Interstitial lung disease (ILD) is an important pulmonary manifestation in pSS patients, referred to as pSS-ILD. The development of ILD in pSS patients marks a major shift in the trajectory of disease development, and the prognosis is often worse compared with pSS patients without lung involvement. pSS-ILD is a complex autoimmune disease characterized by its systemic manifestations and lung involvement, and its treatment usually requires multidisciplinary care including rheumatology, respiratory medicine, and traditional Chinese medicine. Early treatment usually includes glucocorticoids and immunomodulatory drugs. However, the wide geographical distribution of patients, extending to rural areas with limited medical resources, poses a barrier to sustained and effective disease management. This disparity is further exacerbated by the possible underestimation of the importance of regular patient follow-up and high-resolution computed tomography (HRCT), especially in resource-limited settings. These may lead to delayed detection of disease progression. For patients with the fibrotic subtype F-pSS-ILD, the situation is even more severe, and the high cost and out-of-pocket nature of anti-fibrotic treatment often lead to treatment hesitation. This study aims to emphasize the importance of early risk assessment in identifying high-risk pSS-ILD patients who will benefit from enhanced monitoring and regular evaluation, and timely adjustment of treatment regimens are crucial to mitigate disease progression and improve patient survival outcomes.

[0003] pSS-ILD has a series of prognostic risk factors, including but not limited to advanced age, male sex, elevated KL-6 levels, and impaired lung function. Although these clinical features provide insights into the survival prognosis of pSS-ILD patients, it is noteworthy that a comprehensive risk model is lacking to quantitatively evaluate the prognosis of patients. To evaluate its clinical utility and prognostic risk, it is imperative to construct a prediction model that includes a series of risk factors. This model has the potential to identify patients with high mortality or who require lung transplantation at an early stage, allowing for close monitoring and appropriate counseling. Early detection can help to adjust treatment strategies in a timely manner and ultimately prolong life expectancy. Evidence before this study: We searched PubMed and Google Scholar without any language restrictions to identify relevant articles published from January 1, 2000 to May 31, 2023. The search terms were “primary Sjögren's syndrome (pSS)”, “interstitial lung disease (ILD)”, and “prediction model”. Current studies on pSS-ILD have mainly focused on clinical manifestations, risk factors, and prognosis. However, to our knowledge, a risk prediction model for predicting lung transplant-free survival in pSS-ILD patients has not been established.

[0004] The aim of this study was to identify and evaluate the risk factors associated with lung transplant-free survival in patients with pSS-ILD, to establish a prediction model, and to evaluate its accuracy and validity. In addition, clinical decision curve analysis was used to determine the clinical value of the established model. Since lung transplantation (LTx) can be a life-prolonging treatment option for patients with advanced and / or progressive fibrotic interstitial lung disease (ILD), both lung transplantation and death were considered adverse outcomes in our study, that is, the focus was on lung transplant-free survival. This study was funded by the National Key R&D Program, Project No.: 2021YFC2500700. Summary of the invention

[0005] The technical problem to be solved by the present invention is to propose an individualized and visualized nomogram model in view of the deficiencies of the above-mentioned prior art.

[0006] To achieve its purpose, the present invention adopts the following technical scheme: a pSS-ILD lung transplantation-free survival prediction model, characterized in that its construction method comprises the following steps:

[0007] S1. Establish training cohort and external validation cohort;

[0008] S2, perform standard tests on the training cohort and external validation cohort established in step S1, and evaluate them as pSS-ILD through multidisciplinary discussion;

[0009] S3. According to the HRCT pattern, the patients were divided into F-pSS-ILD group and NF-pSS-ILD group and the extent of imaging features, especially the fibrosis score, was quantified.

[0010] S4. Univariate Cox regression analysis was used to evaluate the relationship between clinical characteristics and the prognosis of pSS-ILD, and a nomogram prediction model was developed to predict the 1-, 3-, and 5-year lung transplant-free survival of pSS-ILD patients;

[0011] S5. The prediction accuracy of the nomogram prediction model was evaluated by using the receiver operating characteristic (ROC) curve, the area under the ROC curve, and the consistency index.

[0012] Preferably, in step S1, secondary Sjögren's syndrome-related ILD needs to be excluded.

[0013] Preferably, in step S2, pSS diagnosis refers to the 2002 European and American Consensus Group and the 2016 American College of Rheumatology / European League Against Rheumatism classification standards; the definition of ILD is based on the British Thoracic Society ILD guidelines and the 2002 and 2013 American Thoracic Society / European Respiratory Society idiopathic interstitial pneumonia classification.

[0014] Preferably, in step S3, each HRCT pattern is divided into subgroups of usual interstitial pneumonia, fibrotic nonspecific interstitial pneumonia, cellular NSIP, organizing pneumonia, lymphocytic interstitial pneumonia and follicular bronchiolitis.

[0015] Preferably, in step S4, a total of 4 variables are included in the Cox regression analysis to evaluate independent predictive factors, namely age, oxygenation index, CA125 and fibrosis score.

[0016] Preferably, in step S5, the Hosmer-Lemeshow test and the coefficient of determination are used to evaluate the goodness of fit of the model; the non-parametric bootstrap resampling method is used to draw a calibration curve to evaluate the consistency between the actual observed values ​​and the predicted values ​​of the nomogram; and the decision curve analysis is used to evaluate the clinical utility of the nomogram at different threshold probabilities.

[0017] Further preferably, PASS software is used to evaluate the study sample size, R software is used for statistical analysis, and rms software package is used for regression modeling strategy to provide tools for model validation and diagnostic checks.

[0018] The present invention also aims at the application of the prediction model in the prediction of lung transplant-free survival in pSS-ILD.

[0019] Beneficial effects of the present invention:

[0020] Although the reports on pSS-ILD in the prior art focus on clinical manifestations, prognostic factors and treatment, there is still no model for predicting the probability of death / lung transplantation in pSS-ILD patients. Through the present invention, a risk prediction model based on four factors, namely age, oxygenation index, CA125 and fibrosis score, is constructed. It is an intuitive Nomogram visual model to evaluate the 1-, 3- and 5-year lung transplant-free survival rate, which is of great significance for related research.

[0021] The visual nomogram model provides clinicians with a simple, intuitive and practical prediction tool. It is a pSS-ILD prognosis prediction model based on a large prospective cohort study. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] For ease of explanation, the present invention is described in detail with reference to the following specific embodiments and the accompanying drawings.

[0023] Figure 1 is a flow chart of this embodiment.

[0024] Figure 2 A least absolute shrinkage and selection operator (LASSO) regression feature screening plot with 10-fold cross validation was used for this example.

[0025] Figure 3 This is a nomogram for early prediction of 1-, 3-, and 5-year lung transplant-free survival rates in pSS-ILD patients in this example.

[0026] Figure 4 The ROC curve and AUC graph of the Nomogram of this embodiment are shown in the figure, which show the ROC curves of the 1-, 3-, and 5-year lung transplant-free survival rates of pSS-ILD patients in the training cohort (A) and the external validation cohort (B). ROC: receiver operating characteristic; AUC: area under the ROC curve.

[0027] Figure 5 It is the calibration curve diagram of the Nomogram of the present embodiment, in which the calibration curves of 1, 3, and 5-year lung transplant-free survival of pSS-ILD in the training cohort (A) and the external validation cohort (B). The gray line represents the ideal reference line, in which the predicted probability matches the observed survival rate. The points are calculated by bootstrapping (resampled 1000 times) and represent the performance of the nomogram. The closer the solid line is to the gray line, the higher the accuracy of the model in predicting overall survival. LTx: lung transplant, pSS-ILD: primary Sjögren's syndrome-related interstitial lung disease.

[0028] Figure 6The DCA curve diagram of the Nomogram of this embodiment shows the DCA curves of 1-, 3-, and 5-year lung transplant-free survival of pSS-ILD patients in the training cohort (A, B, C) and the external validation cohort (D, E, F). DCA: decision curve analysis. LTx: lung transplantation, pSS-ILD: primary Sjögren's syndrome-related interstitial lung disease.

[0029] Figure 7 It is the ROC curve diagram of CA125 in this embodiment. DETAILED DESCRIPTION

[0030] The following are specific embodiments of the present invention and the accompanying drawings to further describe the technical solutions of the present invention, but the present invention is not limited to these embodiments; in the following description, specific details such as specific configurations and components are provided only to help fully understand the embodiments of the present invention. Therefore, it should be clear to those skilled in the art that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present invention. In addition, for the sake of clarity and brevity, the description of known functions and structures is omitted.

[0031] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.

[0032] The present invention provides a pSS-ILD lung transplant-free survival prediction model, and the construction method thereof comprises the following steps:

[0033] S1. Establish training cohort and external validation cohort;

[0034] S2, perform standard tests on the training cohort and external validation cohort established in step S1, and evaluate them as pSS-ILD through multidisciplinary discussion;

[0035] S3. Collect basic information such as patient age, gender, course of disease, as well as clinical manifestations such as symptoms and signs, and screen for factors that may affect prognosis. Auxiliary examinations include arterial blood gas analysis, pulmonary function, bronchoalveolar lavage fluid (BALF), tumor markers, etc. For the classification of continuous variables, the ROC curve is used to determine the optimal cutoff value. The optimal cutoff value of CA125 is 35U / ml ( Figure 7 ). Patients were divided into F-pSS-ILD and NF-pSS-ILD groups according to HRCT patterns and the extent of imaging features, especially fibrosis scores, was quantified;

[0036] S4. Univariate Cox regression analysis was used to evaluate the relationship between clinical characteristics and the prognosis of pSS-ILD, and a nomogram prediction model was developed to predict the 1-, 3-, and 5-year lung transplant-free survival of pSS-ILD patients;

[0037] S5. Testing of the nomogram prediction model The receiver operating characteristic (ROC) curve, area under the ROC curve, and consistency index were used to evaluate the prediction accuracy of the model; the nomogram is a useful tool that can provide a visual representation of the predicted probability in the range of 0-100. The sum of each predictor score produces a total score, from which the patient's survival probability can be calculated. The prediction model was externally validated by visualization.

[0038] Preferably, in step S1, secondary Sjögren's syndrome-related ILD needs to be excluded.

[0039] Preferably, in step S2, pSS diagnosis refers to the 2002 European and American Consensus Group and the 2016 American College of Rheumatology / European League Against Rheumatism classification standards; the definition of ILD is based on the British Thoracic Society ILD guidelines and the 2002 and 2013 American Thoracic Society / European Respiratory Society idiopathic interstitial pneumonia classification.

[0040] Preferably, in step S3, each HRCT pattern is then divided into usual interstitial pneumonia (UIP), fibrosing nonspecific interstitial pneumonia (fNSIP), cellular nonspecific interstitial pneumonia (cNSIP), organizing pneumonia (OP), lymphocytic interstitial pneumonia (LIP) and follicular bronchiolitis (FB) subgroups according to the 2002 and 2013 ATS / ERS classifications of IIPs.

[0041] Preferably, in step S4, a total of 4 variables are included in the Cox regression analysis to evaluate independent predictive factors, namely age, oxygenation index, CA125 and fibrosis score.

[0042] Preferably, in step S5, the Hosmer-Lemeshow test and the coefficient of determination are used to evaluate the goodness of fit of the model; the non-parametric bootstrap resampling method is used to draw a calibration curve to evaluate the consistency between the actual observed values ​​and the predicted values ​​of the nomogram; and the decision curve analysis is used to evaluate the clinical utility of the nomogram at different threshold probabilities.

[0043] Further preferably, PASS software is used to evaluate the study sample size, R software is used for statistical analysis, and rms software package is used for regression modeling strategy to provide tools for model validation and diagnostic checks.

[0044] The present invention also aims at the application of the prediction model in the prediction of lung transplant-free survival in pSS-ILD.

[0045] according to Figure 1-7 As shown, this embodiment provides a pSS-ILD lung transplantation-free survival prediction model, and the construction method thereof is as follows:

[0046] After excluding 116 patients, 395 patients were included. Among the 116 excluded patients, 6 had uncontrolled lung infection, 12 had current or previous malignancy, 34 lacked chest HRCT, 15 had follow-up <1 year, and 49 were lost to follow-up. Of the 260 patients in the training cohort, 186 (71.54%) were female, and the median age at diagnosis of pSS-ILD was 64.00 (54.00-69.00) years. The median follow-up time was 2.08 years (1.33-3.60 years). A total of 135 patients were included in the external validation cohort, with a median follow-up time of 4.31 years (2.73-9.99 years). The baseline data of the 260 patients in the training cohort and the 135 patients in the external validation cohort were compared. Comparison of baseline data between the training group and the external validation group showed no significant differences in age, comorbidities, HRCT type, proportion of fibrosis, and imaging characteristics. There are differences in gender, smoking history, disease course, grid score, etc. Detailed procedures and data are available at Figure 1 and Table 1.

[0047] Table 1. Baseline characteristics of all patients in the training cohort and external validation cohort

[0048]

[0049]

[0050] Note: Data are expressed as median (interquartile range) or n (%). Nonparametric tests were used for continuous variables, and χ2 or Fisher's exact test was used for categorical variables. * indicates P < 0.05, indicating statistically significant differences.

[0051] Abbreviations: NF-pSS-ILD: non-fibrotic primary Sjögren's syndrome-related interstitial lung disease; HRCT: high-resolution chest CT; UIP: usual interstitial pneumonia; fNSIP: fibrotic nonspecific interstitial pneumonia; cNSIP: cellular nonspecific interstitial pneumonia; LIP: lymphocytic interstitial pneumonia; OP: organizing pneumonia; FB: follicular bronchiolitis; GGOs: ground-glass opacities.

[0052] Univariate Cox regression analysis showed (Table 2) that age, male sex, smoking history, duration from onset of respiratory symptoms to diagnosis of ILD, duration from diagnosis of pSS to diagnosis of ILD, finger clubbing, Velcro rales, gastroesophageal reflux, pulmonary hypertension, oxygenation index, lung function (FVC% pred, FEV1% pred, DL CO-SB% pred), BALF (lymphocyte% and neutrophil% ), CA125, CA125>35 (U / ml), SSC-Ag, UIP, NSIP, fibrosis, reticular, honeycomb, LIP cystic lesions, reticular score, honeycomb score, and fibrosis score were all statistically significant (P<0.05). The above features were considered as potential predictors. We used LASSO regression based on each feature for feature selection in the training cohort. LASSO analysis retained three non-zero coefficient variables, including oxygenation index, fibrosis score, and CA125>35 (U / ml) ( Figure 2 ). From a clinical perspective, age is significantly associated with prognosis. Therefore, a total of four variables were included in the multivariate Cox regression analysis (Table 2), namely age (HR=1.03, 95%CI 1.00-1.06), oxygenation index (HR=0.99, 95%CI 0.99-0.99), CA125>35 (U / ml) (HR=2.37, 95%CI 1.21-4.61), and fibrosis score (HR=1.93, 95%CI 1.38-2.72). Based on the four factors that affect death / lung transplantation in patients with pSS-ILD, each factor was given a specific score according to its impact on death / lung transplantation. Using the rms package of R software, a visual nomogram was created to represent the prediction model ( Figure 3 ).

[0053] Table 2: Multivariate Cox regression analysis of general characteristics of patients and predictors of screening

[0054]

[0055]

[0056]

[0057] Note: P value was determined by univariate Cox regression analysis, and HR (95% CI) was determined by multivariate Cox regression analysis. * indicates P < 0.05, and the difference is statistically significant.

[0058] Abbreviations: NF-pSS-ILD: nonfibrotic primary Sjögren's syndrome-related interstitial lung disease; MSGB: minor salivary gland biopsy; FVC: forced vital capacity; FEV1: forced expiratory volume in 1 second; DLCO-SB: single-breath diffusion capacity of the lung for carbon monoxide; BALF: bronchoalveolar lavage fluid; NSE: neuron-specific enolase; CEA: carcinoembryonic antigen; CA125: carbohydrate antigen 125; SCC-Ag: squamous cell carcinoma antigen; LTx: lung transplantation; HRCT: high-resolution thoracic CT; UIP: usual interstitial pneumonia; fNSIP: fibrotic nonspecific interstitial pneumonia; cNSIP: cellular nonspecific interstitial pneumonia; LIP: lymphocytic interstitial pneumonia; OP: organizing pneumonia; FB: follicular bronchiolitis; GGOs: ground-glass opacities.

[0059] In the training cohort, the C index was 0.812, and the AUC values ​​at 1, 3, and 5 years were 0.781 (95% CI 0.635-0.927), 0.874 (95% CI 0.804-0.943), and 0.909 (95% CI 0.848-0.969), respectively. In the external validation cohort, the C index at 1, 3, and 5 years was 0.809, and the corresponding AUC values ​​were 0.793 (95% CI 0.662-0.923), 0.826 (95% CI 0.733-0.918), and 0.863 (95% CI 0.788-0.938), respectively. The results showed a sustained level of accuracy ( Figure 4 The calibration curves show that the bias-corrected lines for the training cohort and the external validation cohort overlap closely with the ideal line ( Figure 5 ). DCA illustrates the clinical net benefit that can be achieved at different risk thresholds ( Figure 6 ). These data suggest that the nomogram we constructed has significant potential for clinical decision making. For example, assuming a patient is 55 years old, corresponding to 31 points, oxygenation index (OI) is 450 mmHg (30 points), fibrosis score is 1 point (20 points), CA125 is 40 U / ml (27 points), and the total score is 148 points. This total score corresponds to an estimated 1-, 3-, and 5-year lung transplant-free survival rates of approximately 0.86, 0.40, and 0.16, respectively.

[0060] As we know, age is closely associated with interstitial lung disease, especially with poor prognosis. A study by Everaerts S et al. showed that telomere length progressively decreases with aging of the lung. Elderly patients with pSS-ILD have a higher risk of death / lung transplantation compared with younger patients. Hypoxemia, which refers to abnormally low oxygen levels in the blood, is a key feature of interstitial lung disease. This condition occurs because interstitial lung disease affects the ability of the lungs to effectively exchange gases. Therefore, ILD patients often experience symptoms such as shortness of breath (especially during physical activity) and dry cough. The severity of hypoxemia in ILD patients may vary and often worsens as the disease progresses. Our study showed that the oxygenation index is an independent predictor of the occurrence of death / lung transplantation in patients and was included in the prediction model. The lower the oxygenation index, the higher the nomogram score, and the higher the probability of death / lung transplantation. CA125 is a widely used biomarker for ovarian cancer. It has been cloned and identified as the transmembrane mucin MUC16, which is the largest mucin identified to date and is expressed on the surface of various epithelial cells in the human body. This protein is a marker of epithelial damage. Multiple studies have confirmed that CA125 is elevated in patients with ILD or pSS-ILD. This study specifically found that CA125 was significantly associated with prognosis. Considering that the training cohort and the external validation cohort came from different hospitals and time periods, the CA125 test used the same method and consistent reference interval. The recommended clinical reference value of CA125 is <35.00U / ml, which indicates that pSS-ILD patients with CA125 levels exceeding the clinical reference value have a higher risk of adverse outcomes. It is well known that anti-fibrotic therapy can delay disease progression and improve patient prognosis. Our study also showed that fibrosis was significantly associated with prognosis. In order to quantify more accurately, our cohort specifically evaluated the severity of fibrosis in the form of a score based on HRCT. The results showed that the fibrosis score was positively correlated with the nomogram score, indicating that the higher the fibrosis score, the higher the risk of death / lung transplantation.

[0061] In conclusion, this study evaluated the prognostic factors for death / lung transplantation in patients with pSS-ILD and established a 1-, 3-, and 5-year lung transplant-free survival prediction model. The C index of the training set and the external validation set were 0.812 and 0.809, respectively. The model had good discrimination, calibration, and clinical applicability. The visual nomogram model provided clinicians with a simple, intuitive, and practical prediction tool. This is a pSS-ILD prognostic prediction model based on a large prospective cohort study, which also provides a basic framework for future model optimization in this field.

[0062] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0063] Those skilled in the art to which the present application belongs may make various modifications or additions to the specific embodiments described or replace them in a similar manner, but they will not deviate from the inventive concept of the present application or exceed the scope defined by the attached claims.

Claims

1. A pSS-ILD lung transplant-free survival prediction model, characterized in that: The construction method includes the following steps: S1. Establish training cohort and external validation cohort; S2, perform standard tests on the training cohort and external validation cohort established in step S1, and evaluate them as pSS-ILD through multidisciplinary discussion; S3. According to the HRCT pattern, the patients were divided into F-pSS-ILD group and NF-pSS-ILD group and the extent of imaging features, especially the fibrosis score, was quantified. S4. Univariate Cox regression analysis was used to evaluate the relationship between clinical characteristics and the prognosis of pSS-ILD, and a nomogram prediction model was developed to predict the 1-, 3-, and 5-year lung transplant-free survival of pSS-ILD patients; S5. The prediction accuracy of the nomogram model was evaluated by using the receiver operating characteristic (ROC) curve, the area under the ROC curve, and the consistency index.

2. A pSS-ILD lung transplant-free survival prediction model according to claim 1, characterized in that: In step S1, secondary Sjögren's syndrome-related ILD needs to be excluded.

3. A pSS-ILD lung transplant-free survival prediction model according to claim 1, characterized in that: In step S2, the diagnosis of pSS was based on the 2002 European and American Consensus Group and the 2016 American College of Rheumatology / European League Against Rheumatism classification criteria; the definition of ILD was based on the British Thoracic Society ILD guidelines and the 2002 and 2013 American Thoracic Society / European Respiratory Society idiopathic interstitial pneumonia classifications.

4. A pSS-ILD lung transplant-free survival prediction model according to claim 1, characterized in that: In step S3, each HRCT pattern is divided into usual interstitial pneumonia, fibrotic nonspecific interstitial pneumonia, cellular nonspecific interstitial pneumonia, organizing pneumonia, lymphocytic interstitial pneumonia, and follicular bronchiolitis subgroups.

5. A pSS-ILD lung transplant-free survival prediction model according to claim 1, characterized in that: In step S4, a total of 4 variables were included in the Cox regression analysis to evaluate independent predictive factors, namely age, oxygenation index, CA125, and fibrosis score.

6. A pSS-ILD lung transplant-free survival prediction model according to claim 1, characterized in that: In step S5, the Hosmer-Lemeshow test and determination coefficient were used to evaluate the goodness of fit of the model; the Bootstrap resampling method was used to draw a calibration curve to evaluate the consistency between the actual observed values ​​and the predicted values ​​of the nomogram; and the decision curve analysis was used to evaluate the clinical utility of the nomogram at different threshold probabilities.

7. A pSS-ILD lung transplant-free survival prediction model according to claim 1, characterized in that: PASS software was used to assess the study sample size, R software was used for statistical analysis, and the rms software package was used for regression modeling strategies to provide tools for model validation and diagnostic checks.

8. Use of the prediction model according to any one of claims 1 to 7 in predicting lung transplant-free survival in pSS-ILD.

Images