Validation of Proposed Criteria for Progressive Pulmonary Fibrosis.

Rationale: Criteria for progressive pulmonary fibrosis (PPF) have been proposed, but their prognostic value beyond categorical decline in FVC remains unclear. Objectives: To determine whether proposed PPF criteria predict transplant-free survival (TFS) in patients with non-idiopathic pulmonary fibrosis (IPF) forms of interstitial lung disease (ILD). Methods: A retrospective, multicenter cohort analysis was performed. Patients with diagnoses of fibrotic connective tissue disease-associated ILD, fibrotic hypersensitivity pneumonitis, and non-IPF idiopathic interstitial pneumonia from three U.S. centers and one UK center constituted the test and validation cohorts, respectively. Cox proportional hazards regression was used to test the association between 5-year TFS and ⩾10% FVC decline, followed by 13 additional PPF criteria satisfied in the absence of ⩾10% FVC decline. Measurements and Main Results: One thousand three hundred forty-one patients met the inclusion criteria. A ⩾10% relative FVC decline was the strongest predictor of reduced TFS and showed consistent TFS association across cohorts, ILD subtypes, and treatment groups, resulting in a phenotype that closely resembled IPF. Ten additional PPF criteria satisfied in the absence of 10% relative FVC decline were also associated with reduced TFS in the U.S. test cohort, with 6 maintaining TFS associations in the UK validation cohort. Validated PPF criteria requiring a combination of physiologic, radiologic, and symptomatic worsening performed similarly to their stand-alone components but captured a smaller number of patients. Conclusions: An FVC decline of ⩾10% and six additional PPF criteria satisfied in the absence of such decline identify patients with non-IPF ILD at increased risk for death or lung transplantation.

Proteomic Biomarkers of Survival in Idiopathic Pulmonary Fibrosis.

RATIONALE: Idiopathic pulmonary fibrosis (IPF) causes progressive lung scarring and high mortality. Reliable and accurate prognostic biomarkers are urgently needed. OBJECTIVE: To identify and validate circulating protein biomarkers of IPF survival. METHODS: High-throughput proteomic data were generated using prospectively collected plasma samples from patients with IPF from the Pulmonary Fibrosis Foundation Patient Registry (discovery cohort) and the Universities of California-Davis, Chicago, and Virginia (validation cohort). Proteins associated with three-year transplant-free survival (TFS) were identified using multivariable Cox proportional hazards regression. Those associated with TFS after adjustment for false discovery in the discovery cohort were advanced for testing in the validation cohort, with proteins maintaining TFS association with consistent effect direction considered validated. After combining cohorts, functional analyses were performed, and machine learning used to derive a proteomic signature of TFS. MAIN RESULTS: Of 2921 proteins tested in the discovery cohort (n=871), 231 were associated with differential TFS. Of these, 140 maintained TFS association with consistent effect direction in the validation cohort (n=355). After combining cohorts, validated proteins with strongest TFS association were latent-transforming growth factor beta-binding protein 2 (HR 2.43, 95% CI 2.09-2.82), collagen alpha-1(XXIV) chain (HR 2.21; 95% CI 1.86-2.39) and keratin 19 (HR 1.60; 95% CI 1.47-1.74). In decision curve analysis, a proteomic signature of TFS outperformed a similarly derived clinical prediction model. CONCLUSIONS: In largest proteomic investigation of IPF outcomes performed to date, we identified and validated 140 protein biomarkers of TFS. These results shed important light on potential drivers of IPF progression.

A Deep Learning-Based Radiomic Classifier for Usual Interstitial Pneumonia.

BACKGROUND: Because chest CT scan has largely supplanted surgical lung biopsy for diagnosing most cases of interstitial lung disease (ILD), tools to standardize CT scan interpretation are urgently needed. RESEARCH QUESTION: Does a deep learning (DL)-based classifier for usual interstitial pneumonia (UIP) derived using CT scan features accurately discriminate radiologist-determined visual UIP? STUDY DESIGN AND METHODS: A retrospective cohort study was performed. Chest CT scans acquired in individuals with and without ILD were drawn from a variety of public and private data sources. Using radiologist-determined visual UIP as ground truth, a convolutional neural network was used to learn discrete CT scan features of UIP, with outputs used to predict the likelihood of UIP using a linear support vector machine. Test performance characteristics were assessed in an independent performance cohort and multicenter ILD clinical cohort. Transplant-free survival was compared between UIP classification approaches using the Kaplan-Meier estimator and Cox proportional hazards regression. RESULTS: A total of 2,907 chest CT scans were included in the training (n = 1,934), validation (n = 408), and performance (n = 565) data sets. The prevalence of radiologist-determined visual UIP was 12.4% and 37.1% in the performance and ILD clinical cohorts, respectively. The DL-based UIP classifier predicted visual UIP in the performance cohort with sensitivity and specificity of 93% and 86%, respectively, and in the multicenter ILD clinical cohort with 81% and 77%, respectively. DL-based and visual UIP classification similarly discriminated survival, and outcomes were consistent among cases with positive DL-based UIP classification irrespective of visual classification. INTERPRETATION: A DL-based classifier for UIP demonstrated good test performance across a wide range of UIP prevalence and similarly discriminated survival when compared with radiologist-determined UIP. This automated tool could efficiently screen for UIP in patients undergoing chest CT scan and identify a high-risk phenotype among those with known ILD.

Evaluation of Pulmonary Fibrosis Outcomes by Race and Ethnicity in US Adults.

Importance: Pulmonary fibrosis (PF) is characterized by progressive scarring of lung tissue and poor survival. Racial and ethnic minority populations face the greatest risk of morbidity and mortality from disparities impacting respiratory health, but the pattern of age at clinically relevant outcomes across diverse racial and ethnic populations with PF is unknown. Objective: To compare the age at PF-related outcomes and the heterogeneity in survival patterns among Hispanic, non-Hispanic Black, and non-Hispanic White participants. Design, Setting, and Participants: This cohort study included adult patients with a PF diagnosis and used data from prospective clinical registries: the Pulmonary Fibrosis Foundation Registry (PFFR) for the primary cohort and registries from 4 geographically distinct tertiary hospitals in the US for the external multicenter validation (EMV) cohort. Patients were followed between January 2003 and April 2021. Exposures: Race and ethnicity comparisons between Black, Hispanic, and White participants with PF. Main Outcomes and Measures: Age and sex distribution of participants were measured at the time of study enrollment. All-cause mortality and age at PF diagnosis, hospitalization, lung transplant, and death were assessed in participants over 14 389 person-years. Differences between racial and ethnic groups were compared using Wilcoxon rank sum tests, Bartlett 1-way analysis of variance, and χ2 tests, and crude mortality rates and rate ratios were assessed across racial and ethnic categories using Cox proportional hazards regression models. Results: In total, 4792 participants with PF were assessed (mean [SD] age, 66.1 [11.2] years; 2779 [58.0%] male; 488 [10.2%] Black, 319 [6.7%] Hispanic, and 3985 [83.2%] White); 1904 were in the PFFR and 2888 in the EMV cohort. Black patients with PF were consistently younger than White patients (mean [SD] age at baseline, 57.9 [12.0] vs 68.6 [9.6] years; P < .001). Hispanic and White patients were predominantly male (Hispanic: PFFR, 73 of 124 [58.9%] and EMV, 109 of 195 [55.9%]; and White: PFFR, 1090 of 1675 [65.1%] and EMV, 1373 of 2310 [59.4%]), while Black patients were less likely to be male (PFFR, 32 of 105 [30.5%] and EMV, 102 of 383 [26.6%]). Compared with White patients, Black patients had a lower crude mortality rate ratio (0.57 [95% CI, 0.31-0.97), but for Hispanic patients, the mortality rate ratio was similar to that of White patients (0.89; 95% CI, 0.57-1.35). Mean (SD) hospitalization events per person were highest among Black patients compared with Hispanic and White patients (Black: 3.6 [5.0]; Hispanic, 1.8 [1.4]; and White, 1.7 [1.3]; P < .001). Black patients were consistently younger than Hispanic and White patients at first hospitalization (mean [SD] age: Black, 59.4 [11.7] years; Hispanic, 67.5 [9.8] years; and White, 70.0 [9.3] years; P < .001), lung transplant (Black, 58.6 [8.6] years; Hispanic, 60.5 [6.1] years; and White, 66.9 [6.7] years; P < .001), and death (Black, 68.7 [8.4] years; Hispanic, 72.9 [7.6] years; and White, 73.5 [8.7] years; P < .001). These findings remained consistent in the replication cohort and in sensitivity analyses within prespecified deciles of age groups. Conclusions and Relevance: In this cohort study of participants with PF, racial and ethnic disparities, especially among Black patients, were found in PF-related outcomes, including earlier onset of death. Further research is essential to identify and mitigate the underlying responsible factors.