Interstitial Lung Disease Patterns Research Articles

Abatacept in interstitial lung disease associated with rheumatoid arthritis: national multicenter study of 263 patients.

To assess the efficacy of abatacept (ABA) in RA patients with interstitial lung disease (ILD) (RA-ILD). This was an observational, multicentre study of RA-ILD patients treated with at least one dose of ABA. ILD was diagnosed by high-resolution CT (HRCT). We analysed the following variables at baseline (ABA initiation), 12 months and at the end of the follow-up: Modified Medical Research Council (MMRC) scale (1-point change), forced vital capacity (FVC) or diffusion lung capacity for carbon monoxide (DLCO) (improvement or worsening ≥10%), HRCT, DAS on 28 joints evaluated using the ESR (DAS28ESR) and CS-sparing effect. We studied 263 RA-ILD patients [150 women/113 men; mean (s.d.) age 64.6 (10) years]. At baseline, they had a median duration of ILD of 1 (interquartile range 0.25-3.44) years, moderate or severe degree of dyspnoea (MMRC grade 2, 3 or 4) (40.3%), FVC (% of the predicted) mean (s.d.) 85.9 (21.8)%, DLCO (% of the predicted) 65.7 (18.3) and DAS28ESR 4.5 (1.5). The ILD patterns were: usual interstitial pneumonia (UIP) (40.3%), non-specific interstitial pneumonia (NSIP) (31.9%) and others (27.8%). ABA was prescribed at standard dose, i.v. (25.5%) or s.c. (74.5%). After a median follow-up of 12 (6-36) months the following variables did not show worsening: dyspnoea (MMRC) (91.9%); FVC (87.7%); DLCO (90.6%); and chest HRCT (76.6%). A significant improvement of DAS28ESR from 4.5 (1.5) to 3.1 (1.3) at the end of follow-up (P < 0.001) and a CS-sparing effect from a median 7.5 (5-10) to 5 (2.5-7.5) mg/day at the end of follow-up (P < 0.001) was also observed. ABA was withdrawn in 62 (23.6%) patients due to adverse events (n = 30), articular inefficacy (n = 27), ILD worsening (n = 3) and other causes (n = 2). ABA may be an effective and safe treatment for patients with RA-ILD.

CLINICAL CHARACTERISTICS, TREATMENT RESPONSE, AND SURVIVAL OF PATIENTS WITH IDIOPATHIC INFLAMMATORY MYOSITIS AND PULMONARY ARTERIAL HYPERTENSION

CLINICAL CHARACTERISTICS, TREATMENT RESPONSE, AND SURVIVAL OF PATIENTS WITH IDIOPATHIC INFLAMMATORY MYOSITIS AND PULMONARY ARTERIAL HYPERTENSION

Radiological Patterns of Drug-induced Interstitial Lung Disease (DILD) in Early-phase Oncology Clinical Trials.

Drug-induced interstitial lung disease (DILD) is a rare, but potentially fatal toxicity. Clinical and radiological features of DILD in the early experimental setting are poorly described. A total of 2,499 consecutive patients with advanced cancer on phase I clinical trials were included. DILD was identified by a dedicated radiologist and investigators, categorized per internationally recognized radiological patterns, and graded per Common Terminology Criteria for Adverse Events (CTCAE) and the Royal Marsden Hospital (RMH) DILD score. Clinical and radiological features of DILD were analyzed. Sixty patients overall (2.4%) developed DILD. Median time to onset of DILD was 63 days (range, 14-336 days). A total of 45% of patients who developed DILD were clinically asymptomatic. Incidence was highest in patients receiving drug conjugates (7.4%), followed by inhibitors of the PI3K/AKT/mTOR pathway (3.9%). The most common pattern seen was hypersensitivity pneumonitis (33.3%), followed by nonspecific interstitial pneumonia (30%), and cryptogenic organizing pneumonia (26.7%). A higher DILD score [OR, 1.47, 95% confidence interval (CI), 1.19-1.81; P < 0.001] and the pattern of DILD (OR, 5.83 for acute interstitial pneumonia; 95% CI, 0.38-90.26; P = 0.002) were significantly associated with a higher CTCAE grading. The only predictive factor for an improvement in DILD was an interruption of treatment (OR, 0.05; 95% CI, 0.01-0.35; P = 0.01). DILD in early-phase clinical trials is a toxicity of variable onset, with diverse clinical and radiological findings. Radiological findings precede clinical symptoms. The extent of the affected lung parenchyma, scored by the RMH DILD score, correlates with clinical presentation. Most events are low grade, and improve with treatment interruption, which should be considered early.

Interstitial Lung Disease in Giant Cell Arteritis: Review of 23 Patients.

Giant cell arteritis (GCA) is a large-vessel vasculitis with systemic manifestations. A few case reports have described a possible association of GCA with interstitial lung disease (ILD). The primary aim of the present study was to describe the pattern, severity, and course of ILD in patients with GCA. This medical records review study evaluated adult patients presenting to Mayo Clinic in Rochester, MN, from January 1, 1997, through December 31, 2018, who had the diagnoses of GCA and ILD. Clinical, laboratory, and radiologic data were analyzed. In total, 23 patients were in the study. Median (range) age was 78 (58-93) years, and 14 (61%) were women. Six patients (26%) had a cough at GCA diagnosis. At ILD diagnosis, 15 patients had respiratory symptoms, including dyspnea (n = 12, 52%), dry cough (n = 6, 26%), wheezing (n = 1, 4%), and chest pain (n = 1, 4%). On initial chest computed tomography, the most common pattern of ILD was probable usual interstitial pneumonia (n = 7, 30%), indeterminate for usual interstitial pneumonia (n = 5, 22%), and combined pulmonary fibrosis and emphysema (n = 3, 13%). Airway abnormalities were present in 10 patients: 6 with bronchial wall thickening, 2 with bronchiectasis, and 2 with both. At follow-up computed tomography, 8 patients had ILD progression. Three patients with cough improved after initiation of glucocorticoid therapy. Interstitial lung disease and airway abnormalities may be associated with GCA. Although cough may improve, ILD in some patients with GCA may progress despite immunosuppressive therapy.

Is exposure to environmental factors associated with a characteristic clinical and laboratory profile in systemic sclerosis? A retrospective analysis.

To identify environmental factors (EF) in a large cohort of patients with systemic sclerosis (SSc) analyzing their clinical and laboratory presentation. A cohort of consecutive patients attended at a single Brazilian SSc outpatient clinic was analyzed regarding EF. Data were analyzed according to clinical, demographic and laboratory characteristics, as well as SSc subtype. In a cohort of 662 patients, 70 (10.6%) had known previous exposure to EF, predominantly organic solvents (51.4%), silica (20%), silicone (12.9%) and pesticides (11.4%). In the SSc cohort, patients with EF had a significantly higher frequency of male gender (p < 0.01), African-Brazilian ethnicity (p = 0.01), myopathy (p = 0.02), and pigmentary disorders (p = 0.04), with shorter disease duration (p = 0.01). When SSc subtypes were analyzed separately, there was positive association with male gender in limited (p < 0.01) and diffuse (p < 0.01) SSc, as well as African-Brazilian ethnicity (p = 0.04), severe interstitial lung disease (p < 0.01), myopathy (p = 0.02) and SD pattern at nailfold capillaroscopy (p = 0.01) in limited SSc, and negative association with esophageal hypomotility (p < 0.01) and ANA positivity (p = 0.02) in diffuse SSc. Multiple regression analyses showed that myopathy was independently associated with previous exposure to EF (OR = 2.09; 95% CI 1.15-3.82), especially silica exposure (OR = 3.09; 95% CI 1.67-5.73). This study showed that SSc patients with previous exposure to EF may have some specific clinical characteristics, mainly a higher frequency of myopathy, also showing more severe ILD, preferably in male and African-Brazilian patients, associated with a lower frequency of ANA positivity.

Prevalence and clinical correlates of small airway obstruction in patients with systemic sclerosis.

This study aims to assess the prevalence and clinical correlates of small airway obstruction (SAO) in patients with systemic sclerosis (SSc). Sixty-nine consecutive patients with SSc (63 women and 6 men) were included. Lung function tests, including assessment of lung diffusing capacity, were performed in all patients. Patients were considered to have SAO when the maximal expiratory flow at 25% of the forced vital capacity (MEF25) was lower than 60% as predicted. High-resolution computed tomography (HRCT) of the lung was performed in all patients with MEF25 < 60%. We assessed the relationship of SAO in our patients with large airway obstruction, decreased lung diffusing capacity, HRCT findings, disease duration, disease subtype, scleroderma-specific antibodies, and smoking. SAO was noticed in 46/69 (66.6%) of patients with SSc. Restrictive lung disease was found in 4/69 (5.8%), obstruction of large airways in 18/69 (26.1%), and decreased lung diffusing capacity in 47/69 (68.1%) of patients. No difference in gender, age, disease duration, disease subtype, and scleroderma-specific antibodies was found between patients with and without SAO. Eighteen out of forty-six (39.1%) patients with SAO had decreased forced expiratory volume in 1 sec (FEV1) and the Tiffeneau-Pinelli index, indicating presence of coexistent large airway obstruction. Twenty out of forty-six (43.5%) patients with SAO had associated decreased lung diffusing capacity, while 8/46 (17.4%) of patients had isolated SAO. HRCT patterns of interstitial lung disease (ILD) were found more frequently in patients with SAO and decreased lung diffusing capacity, compared with patients with SAO and normal diffusing capacity (75% vs 11.5%, p = 0.008). We have noticed that tobacco smokers among SSc patients with SAO have more common associated obstructive lung disease on spirometry (58.8% vs 15.4%, p = 0.004). On the other hand, isolated SAO and SAO associated with impaired diffusing capacity were equally frequent among smokers and non-smokers. Patients with SSc have commonly SAO. It can be considered clinical feature of undiagnosed asthma or chronic obstructive pulmonary disease (COPD), if isolated or associated with large airway obstruction, especially in tobacco smokers. On the other hand, SAO associated with decreased lung diffusing capacity was found to be not related to smoking, and may indicate a possible prominent bronchiolar involvement within SSc-related interstitial lung disease Key Points • Small airway obstruction in patients with systemic sclerosis can be considered a clinical feature of undiagnosed obstructive lung disease, if isolated or associated with large airway obstruction, especially in tobacco smokers. • Obstruction of small airways, associated with decreased lung diffusing capacity, may indicate a possible prominent bronchiolar involvement within systemic sclerosis-related interstitial lung disease.

Clinical and radiological spectrum of interstitial lung disease in a north Indian state: A single centre study

Background: The present study was carried out with an aim of understanding the types of Interstitial lung disease (ILD) and their radiological patterns in the north Indian state of Jammu and Kashmir. Methods: The present study was a prospective observational study conducted over a period of 2 years in Sher-i-Kashmir Institute of Medical Sciences- Medical College, Bemina, Srinagar. All consecutive patients referred to the department for High Resolution CT scan (HRCT) of the chest with suspected diagnosis of ILD based on clinical features, spirometry and chest X-ray findings, who gave consent for the study were included. Clinical, biochemical, serological and radiological parameters were assessed and the diagnosis of various ILDs was made based on standard international criteria. Results: A total 122 patients of ILD were enrolled over a period of 2 years in the present study. Mean age of the patients was 63.4 ± 19.7 years and majority (59.8%) were males. 41.8% patients were current or previous smokers. Mean symptom duration before diagnosis was 3.1 ± 1.7 years. Exertional breathlessness (98.4%) and cough (94.3%) were the most common symptoms. Hypertension (70.5%) and gastroesophageal reflux disease (62.3%) were the most frequent comorbidities. On HRCT thorax ground glass opacities (52.5%) were the most common finding, followed by interlobular (39.3%) and intralobular (34.4%) septal thickening. Diffuse pattern of lung involvement was seen in 41.8% and predominant upper/middle lobe involvement in 33.6%. Idiopathic pulmonary fibrosis (29.5%) was the most common form of ILD, followed by Nonspecific interstitial pneumonia (23.8%) and hypersensitivity pneumonitis (15.6%). Gujjar lung, a unique form of ILD in Kashmir, was diagnosed in 2.5% patients. Conclusion: The present study is the first from the state of Jammu and Kashmir assessing the radiological spectrum and types of ILD in this part of country. It provides a useful insight into the clinical and radiological patterns of ILD in our population and highlights the need for larger prospective multicentre registries to assess the true incidence and prevalence of ILD in the country.

Quantitative analysis of pulmonary vasculature in systemic sclerosis at spirometry-gated chest CT

ObjectiveTo prospectively investigate whether differences in pulmonary vasculature exist in systemic sclerosis (SSc) and how they are distributed in patients with different pulmonary function.MethodsSeventy-four patients with SSc undergoing chest CT...

Interaction between the promoter MUC5B polymorphism and mucin expression: is there a difference according to ILD subtype?

The MUC5B promoter variant rs35705950 is associated with idiopathic pulmonary fibrosis (IPF). MUC5B glycoprotein is overexpressed in IPF lungs. We examined immunohistochemical expression of MUC5B in different interstitial lung disease...

SAT0553 QUANTITATIVE ANALYSIS OF IMAGING FEATURES AT CHEST CT OF PULMONARY ARTERIAL AND VENOUS COMPONENTS IN SYSTEMIC SCLEROSIS-INTERSTITIAL LUNG DISEASE (SSc-ILD).

Background:Interstitial lung disease (ILD) and pulmonary arterial hypertension carry a negative impact on SSc prognosis. Chest CT is the gold standard in assessing ILD and helps in evaluating associated vascular involvement.Objectives:As qualitative analysis of CT scans is limited by low reproducibility and time constraints, we aimed at evaluating parenchymal and vascular features in SSc-ILD by quantitative analysis (QA) of CT scans and testing the relationship with clinical-functional data.Methods:We prospectively enrolled 80 patients who underwent PFTs and chest CT scan spirometry gated at TLC on the same day. Clinical, lung functional and diffusion data, as well as disability indexes were collected. CT images were analyzed by a computational platform for texture analysis of ILD patterns (CALIPER), through Imbio LTA. It quantified the extent of normal pattern (NP %), ground glass opacities (GG %), reticulation (RET %), honeycombing (HC %), total ILD extent (ILD EXT %) and hyperlucent (HL %). Low density areas, representing emphysematous area, were also quantified (LDA %). For lung vessel analysis, a software program developed by the Ludwig Boltzmann Institute for Lung Vascular Research was used. This software determined total, arterial, and venous vascular volumes (TV, AV, VV), and relative volumes (TV%, AV%, VV%), as well as density and number for total, arterial and venous vessels.Results:43/80 patients/CT scans were eligible for both software analyses, while 36/43 for arterial and venous separation. TV% and total vessel density were correlated positively with mRSS and negatively with %FVC (r=-0.537 and r=-0.382) and %TLC (r=-0.511 and r=-0.648), while vessel tortuosity correlated positively with %DLco. This was confirmed when separately analyzing arterial vessels, while VV% negatively correlated with %FVC, %TLC and %DLco. There was a positive correlation between %ILD patterns and %vascular volumes, being significant for TV%-AV%, total vessels and arterial density. Conversely, %ILD patterns were negatively correlated with VV and number of veins detected, despite positive correlation between VV% and ILD_EXT%. When clustering patients according to %FVC and %DLco with 80% normal cutoff, %FVC allowed clustering according to significantly different ILD patterns extents and vascular features, while %DLCO for vascular features only. Moreover, the consecutive addition of functional impairment and worsening of ILD (from both normal %FVC and %Dlco, to %DLco impairment only to both %FVC and %Dlco impairment), there was a significant increase in %TV, % AV and %VV, with the exception of decrease in %VV and venous density in patients with double impairment versus DLco single impairment.Conclusion:This is the first study showing in SSc a direct correlation between ILD and the increase in lung vascular volume, which is characterized by increase in arterial volume and density and reduction in venous volume and number. These results might be explained by the reduction of pulmonary volume due to fibrosis. However, also a para-physiological mechanism of redistribution of blood flow in lung areas, less involved by ILD, might be considered. Further studies on lung vessel quantification and distribution are ongoing.Disclosure of Interests:Cosimo Bruni Speakers bureau: Actelion, Eli Lilly, Mariaelena Occhipinti Consultant of: Imbio, Gianna Camiciottoli: None declared, Maurizio Bartolucci: None declared, Michael Pienn: None declared, Gemma Lepri: None declared, Alessio Fabbrizzi: None declared, Alessandra Tottoli: None declared, Giuglia Ciardi: None declared, Dilia Giuggioli: None declared, Giovanna CUOMO: None declared, Francesco Masini: None declared, Horst Olschewski: None declared, Federico Lavorini: None declared, Linda Calistri: None declared, Marco Matucci-Cerinic Grant/research support from: Actelion, MSD, Bristol-Myers Squibb, Speakers bureau: Acetelion, Lilly, Boehringer Ingelheim

OP0181 FUNCTIONAL CUT-OFFS TO DISTINGUISH PULMONARY VASCULAR AND PARENCHYMAL INVOLVEMENT IN SYSTEMIC SCLEROSIS (SSC): A QUANTITATIVE ANALYSIS OF IMAGING FEATURES AT CHEST COMPUTED TOMOGRAPHY (CT)

Background:Interstitial lung disease (ILD) and pulmonary arterial hypertension represent the most frequent causes of morbidity and mortality in SSc, with chest CT representing the gold standard in ILD assessment, while FVC and DLco allow functional assessment.Objectives:As qualitative analysis of given chest CT scans is hampered by low reproducibility, we aimed to perform a quantitative analysis (QA) of CT scans able to investigate the parenchymal and vascular features in SSc-ILD and thus testing the relationship with clinical-functional data.Methods:We prospectively enrolled 80 patients who underwent PFTs and spirometry-gated chest CT scan at TLC on the same day. Clinical, lung functional and diffusion data, as well as disability indexes were collected. CT images were analyzed by a computational platform for texture analysis of ILD patterns (CALIPER) through Imbio LTA. It quantified the extent of normal lung (%N), ground-glass opacities (%GG), reticulation (%RET), honeycombing (%HC), hyperlucent (%HL), absolute (PVV, cm3) and normalized (PVV/LV, %) pulmonary vascular volumes. Cut-offs of normality for %FVC and %DLco of 80% and 70% were tested to differentiate parenchymal and vascular features.Results:73 patients/CT scans were eligible for both software analyses. CALIPER showed GG% as the most frequent radiological pattern (mean 5.5±10.4%). %FVC and % TLC negatively correlated with all ILD patterns, while %DLco with RET% only; PVV and PVV/LV negatively correlated with %FVC and %TLC, while %DLco with PVV/LV only. Positive correlations were found between all ILD patterns and vascular volumes (Table 1).LV (cm3)%N%GG%RET%HC%HLPVV (cm3)% PVV/LVFVC%r.60-.19-.40-.34-.30.35-.26-.44p&lt;.001-&lt;.001.004.01.003.04&lt;.001FEV1%r.58-.02-.38-.25-.24.23-.35-.49p&lt;.001-.002.04.05-.004&lt;.001FEV1/FVCr-.16.33.22.16.21-.35-.15-.08p-.02------TLC%r.71-.14-.42-.37-.48.40-.43-.64p&lt;.001-.001.01&lt;.001.002&lt;.001&lt;.02DLco%r.38-.05-.21-.31-.22.30-.21-.33p.01--.01---.006FVC/DLcor.03-.08-.06-.003-.09.08-.06-.08p--------Cut-offs equal to 80 for %FVC and 70 for %DLco distinguished both parenchymal and vascular features, while 80 for %DLco characterized vascular features only. These results were confirmed also when patients were stratified according to absent/single/combined %FVC and %DLCO impairments with 80% cut-offs (Table 2).FVC&lt;80%FVC ≥80%pDLco&lt;80%DLco ≥80%pDLco &lt;70%DLco ≥70%p%N82.7 (9.6)86.2 (14.7)-86.6 (12.7)80.8 (15.8)-84.1 (13.9)86.4 (13.5)-%GG10.3 (8.9)2.4 (3.9)&lt;.0015.0 (6.7)3.9 (6.9)-6.2 (7.5)2.4 (4.8).002%RET2.9 (2.9)0.8 (1.3)&lt;.0011.6 (2.1)0.7 (0.9)-1.9 (2.4)0.6 (0.8).007%HC0.4 (0.6)0.1 (0.1)&lt;.0010.2 (0.3)0.1 (0.1)-0.2 (0.4)0.05 (0.2).010%HL3.6 (6.8)8.9 (12.1)-5.4 (8.8)14.1 (15.4).0506.3 (10.1)9.2 (12.7)-PVV125.6 (39.1)90.9 (26.9)&lt;.001101.9 (34.8)84.7 (19.4).016106.9 (38.3)87.5 (20.5).012PVV/LV3.8 (1.6)2.0 (0.7)&lt;.0012.51 (1.3)1.7 (0.6).0022.76 (1.4)1.83 (0.6).001Conclusion:In SSc a cut-off at 80 for %DLco may help identifying vascular changes as automatically assessed on chest CT scan, without any underlying ILD. The 80% cut-off for %DLco may be proposed to identify isolated vascular involvement, while %FVC at 80% or %DLco at 70% to identify significant parenchymal involvement. These results need to be confirmed in larger multi-centric cohorts.Disclosure of Interests:Cosimo Bruni Speakers bureau: Actelion, Eli Lilly, Mariaelena Occhipinti Consultant of: Imbio, Gianna Camiciottoli: None declared, Maurizio Bartolucci: None declared, Gemma Lepri: None declared, Alessio Fabbrizzi: None declared, Alessandra Tottoli: None declared, Anna Bassetto: None declared, Giuglia Ciardi: None declared, Dilia Giuggioli: None declared, Giovanna CUOMO: None declared, Francesco Masini: None declared, Federico Lavorini: None declared, Linda Calistri: None declared, Marco Matucci-Cerinic Grant/research support from: Actelion, MSD, Bristol-Myers Squibb, Speakers bureau: Acetelion, Lilly, Boehringer Ingelheim

HRCT Patterns of Drug-Induced Interstitial Lung Diseases: A Review.

Interstitial Lung Diseases (ILDs) represent a heterogeneous group of pathologies, which may be related to different causes. A low percentage of these lung diseases may be secondary to the administration of drugs or substances. Through the PubMed database, an extensive search was performed in the fields of drug toxicity and interstitial lung disease. We have evaluated the different classes of drugs associated with pulmonary toxicity. Several different high resolution computed tomography (HRCT) patterns related to pulmonary drug toxicity have been reported in literature, and the most frequent ILDs patterns reported include Nonspecific Interstitial Pneumonia (NSIP), Usual Interstitial Pneumonia (UIP), Hypersensitivity Pneumonitis (HP), Organizing Pneumonia (OP), Acute Respiratory Distress Syndrome (ARDS), and Diffuse Alveolar Damage (DAD). Finally, from the electronic database of our Institute we have selected and commented on some cases of drug-induced lung diseases related to the administration of common drugs. As the imaging patterns are rarely specific, an accurate evaluation of the clinical history is required and a multidisciplinary approach—involving pneumologists, cardiologists, radiologists, pathologists, and rheumatologists—is recommended.

Patients with Interstitial Lung Disease Secondary to Autoimmune Diseases: How to Recognize Them?

The diagnostic assessment of patients with Interstitial Lung Disease (ILD) can be challenging due to the large number of possible causes. Moreover, the diagnostic approach can be limited by the severity of the disease, which may not allow invasive exams. To overcome this issue, the referral centers for ILD organized Multidisciplinary Teams (MDTs), including physicians and experts in complementary discipline, to discuss the management of doubtful cases of ILD. MDT is currently considered the gold standard for ILD diagnosis, but it is not often simple to organize and, furthermore, rheumatologists are still not always included. In fact, even if rheumatologic conditions represent a common cause of ILD, they are sometimes difficult to recognize, considering the variegated clinical features and their association with all possible radiographic patterns of ILD. The first objective of this review is to describe the clinical, laboratory, and instrumental tests that can drive a diagnosis toward a possible rheumatic disease. The secondary objective is to propose a set of first-line tests to perform in all patients in order to recognize any possible rheumatic conditions underlying ILD.

Differential diagnosis of Interstitial Lung Diseases using Deep Learning networks

ABSTRACT An architecture for automatic lung tissue classification method based on the Deep Learning techniques is designed in this paper. Recent works on Deep Learning techniques achieved impressive results in the field of medical image classification. So, we designed a Convolution Neural Network (CNN) for the classification of five categories of Interstitial Lung Diseases (ILD) patterns in High-Resolution Computed Tomography (HRCT) images. The CNN consists of 3 Convolution layers, Leaky ReLU activation followed by Maximum pooling layer and dense layer. The last Fully Connected (FC) layer has 5 outputs equivalent to the classes considered such as Normal, Ground Glass (GG), Emphysema, Micro Nodules, and Fibrosis. The proposed CNN is trained and evaluated on the publicly available ILD database provided by the University Hospitals of Geneva (HUG). Experimental results are compared with the state-of-art, which shows an outstanding performance of the proposed CNN model giving 94.67% accuracy and 94.65% Favg .

Prognostic significance of the radiologic features of pneumonitis induced by anti-PD-1 therapy.

BackgroundInterstitial lung disease (ILD) induced by anti‐programmed‐cell death‐1 (PD‐1) and anti‐PD‐ligand 1 (PD‐L1) is potentially life‐threatening and is a common reason of the discontinuation of therapy. In contrast, an enhancement in antitumor effects was reported in patients who developed immune‐related adverse events, including ILD. Although recent evidence suggests that radiologic patterns of ILD may reflect the severity of ILD and the antitumor immune responses to anti‐PD‐1/PD‐L1 therapies, the association between radiologic features and clinical outcomes remains unclear.MethodsPatients with advanced non‐small‐cell lung cancer who were treated with 1st to 3rd line anti‐PD‐1 therapy from January 2016 through October 2017 were identified at multiple institutions belonging to the Niigata Lung Cancer Treatment Group. ILD was diagnosed by the treating physicians, and chest computed tomography scans were independently reviewed to assess the radiologic features of ILD.ResultsA total of 231 patients who received anti‐PD‐1 therapy were enrolled. Thirty‐one patients (14%) developed ILD. Sixteen patients were classified as having ground glass opacities (GGO), 16 were classified as having cryptogenic organizing pneumonia (COP), and one was classified as having pneumonitis not otherwise specified. Patients with GGO had significantly worse overall survival time compared to patients with COP (7.8 months (95% CI: 2.2‐NE) versus not reached (95% CI: 13.2‐NE); P = 0.0175). Multivariate analysis of all 231 patients also revealed that PS = 1 and ≥2 and GGO were significant predictors of a worse overall survival.ConclusionsThis study demonstrated that patients who developed GGO exhibited worse outcomes among non‐small‐cell lung cancer patients receiving anti‐PD‐1 therapies.

Deep learning for screening of interstitial lung disease patterns in high-resolution CT images

To develop a screening tool for the detection of interstitial lung disease (ILD) patterns using a deep-learning method. A fully convolutional network was used for semantic segmentation of several ILD patterns. Improved segmentation of ILD patterns was achieved using multi-scale feature extraction. Dilated convolution was used to maintain the resolution of feature maps and to enlarge the receptive field. The proposed method was evaluated on a publicly available ILD database (MedGIFT) and a private clinical research database. Several metrics, such as success rate, sensitivity, and false positives per section were used for quantitative evaluation of the proposed method. Sections with fibrosis and emphysema were detected with a similar success rate and sensitivity for both databases but the performance of detection was lower for consolidation compared to fibrosis and emphysema. Automatic identification of ILD patterns in a high-resolution computed tomography (CT) image was implemented using a deep-learning framework. Creation of a pre-trained model with natural images and subsequent transfer learning using a particular database gives acceptable results.

A deep convolutional neural network architecture for interstitial lung disease pattern classification.

Interstitial lung disease (ILD) refers to a group of various abnormal inflammations of lung tissues and early diagnosis of these disease patterns is crucial for the treatment. Yet it is difficult to make an accurate diagnosis due to the similarity among the clinical manifestations of these diseases. In order to assist the radiologists, computer-aided diagnosis systems have been developed. Besides, the potential of deep convolutional neural networks (CNNs) is also expected to exert on the medical image analysis in recent years. In this paper, we design a new deep convolutional neural network (CNN) architecture to achieve the classification task of ILD patterns. Furthermore, we also propose a novel two-stage transfer learning (TSTL) method to deal with the problem of the lack of training data, which leverages the knowledge learned from sufficient textural source data and auxiliary unlabeled lung CT data to the target domain. We adopt the unsupervised manner to learn the unlabeled data, by which the objective function composed of the prediction confidence and mutual information are optimized. The experimental results show that our proposed CNN architecture achieves desirable performance and outperforms most of the state-of-the-art ones. The comparative analysis demonstrates the promising feasibility and advantages of the proposed two-stage transfer learning strategy as well as the potential of the knowledge learning from lung CT data. Graphical Abstract The framework of the proposed two-stage transfer learning method.

Classification of Interstitial Lung Abnormality Patterns with an Ensemble of Deep Convolutional Neural Networks

Subtle interstitial changes in the lung parenchyma of smokers, known as Interstitial Lung Abnormalities (ILA), have been associated with clinical outcomes, including mortality, even in the absence of Interstitial Lung Disease (ILD). Although several methods have been proposed for the automatic identification of more advanced Interstitial Lung Disease (ILD) patterns, few have tackled ILA, which likely precedes the development ILD in some cases. In this context, we propose a novel methodology for automated identification and classification of ILA patterns in computed tomography (CT) images. The proposed method is an ensemble of deep convolutional neural networks (CNNs) that detect more discriminative features by incorporating two, two-and-a-half and three- dimensional architectures, thereby enabling more accurate classification. This technique is implemented by first training each individual CNN, and then combining its output responses to form the overall ensemble output. To train and test the system we used 37424 radiographic tissue samples corresponding to eight different parenchymal feature classes from 208 CT scans. The resulting ensemble performance including an average sensitivity of 91,41% and average specificity of 98,18% suggests it is potentially a viable method to identify radiographic patterns that precede the development of ILD.

Picking interstitial lung disease out of the myositis haystack

Interstitial lung disease (ILD) is a common manifestation of the connective tissue disease (CTD) associated idiopathic inflammatory myopathies (IIM). Although patients may be diagnosed as having polymyositis (PM) or dermatomyositis (DM) under the IIM spectrum, it is quite clear that risk, pattern of ILD and disease course between subgroups of patients is different. The natural history may be asymptomatic and slowly progressive or stable, chronically progressive or fulminant rapidly progressive depending on ILD subtype. ILD can be the initial presenting feature and this can make recognition of an underlying CTD-IIM overlap more difficult with some patients initially misdiagnosed with idiopathic pulmonary fibrosis. Therefore, early recognition and characterization of patients can influence management and prognosis. It is clear than certain clinical and serological features phenotype patients into more specific CTD-IIM ILD subgroups. A number of myositis-CTD overlap associated antibodies and their clinical patterns have been described over the last few years. The hallmark CTD-IIM ILD subgroup is antisynthetase syndrome, characterized by autoantibodies to tRNA synthetases. Muscle weakness is not universally present and parenchymal lung disease can predominate. Anti-MDA5 DM has a distinct cutaneous pulmonary phenotype and is significantly associated with the development of ILD with different patterns seen in different ethnic groups. Other autoantibodies associated with ILD include those targeting nucleolar autoantigens such as anti-PM-SCL, again with characteristic syndromes. Picking ILD out of the “myositis haystack” can be complex. This heterogeneous disease group requires robust multidisciplinary collaboration between rheumatologists, pulmonologists, thoracic radiologists, and histopathologists to bring together clinical assessment to reach a diagnostic conclusion so optimal outcomes can be achieved.

Is operation safe for lung cancer patients with interstitial lung disease on computed tomography?

Aims:Interstitial lung disease (ILD) is associated with the incidence of non-small cell lung cancer (NSCLC). Patients with ILD are at risk of acute exacerbation (AE) after pulmonary resection. However, there have been no recognized treatment guidelines for NSCLC patients with ILD on computed tomography (CT).Methods:We reviewed the medical records of 156 consecutive patients with ILD on high-resolution CT who have undergone pulmonary resection and between 2014 and 2018. Data regarding general information, imaging features, perioperative indicators, and long-term prognosis of patients were compared.Results:The mean patient age was 67.24 ± 6.80 years. Postoperative AE occurred in seven (4.5%) patients; five (71.4%) of the seven patients who had an AE died within 30 days. The incidence of postoperative AE was 5.3% among patients who underwent lobectomy (n = 6). Overall survivals (OS) was significantly poorer in patients with possible usual interstitial pneumonia (UIP)/UIP [hazard ratio (HR) 2.34, 95% confidence interval (CI) 1.11–4.95, p = 0.026] and severe postoperative complications (Grade ⩾3) (versus no complication: HR 2.58, 95% CI 1.11–6.02, p = 0.028; versus mild complications: HR 6.05, 95% CI 2.69–13.6, p < 0.001). Age (HR 1.071, 95% CI 1.006–1.137, p = 0.030) and ILD patterns (HR 2.420, 95% CI 1.024–5.716, p = 0.044) were independent prognostic factors for OS. Forced vital capacity (FVC) (odds ratio 0.351, 95% CI 0.145–0.850, p = 0.020) was an independent prognostic factor for patients who needed postoperative intensive care unit intervention.Conclusion:Pulmonary resection for NSCLC Patients with ILD on CT is a safe procedure. However, surgical indications for lobectomy need to be more carefully for these patients, especially for possible UIP/UIP patients and patients with lower FVC. The reviews of this paper are available via the supplemental material section.