Citations

Abstract RATIONALE: Research since 2017 has highlighted artificial intelligence (AI)-based technologies to assist in detecting and/or classifying lung nodules on Computed Tomography (CT) scans. However, few have combined detection and classification tasks, and most have required extensive hands-on image manipulation to process the cases, which is both costly and time-consuming. Here, we present a closed loop, fully automated software package, Bronchosolve, which handles cases from scan to result without manual intervention. METHODS: The software package combined pre-processing, analysis, and result generation system including a deep learning convolutional neural network (CNN) for assessment of suspicious pulmonary nodules. Model inputs were full CT cases formatted in Digital Imaging and Communications in Medicine (DICOM), without demographics or clinical variables. Automated steps included 1) automatic selection of the optimal CT series for analysis; 2) normalization and pre-processing of the series; 3) AI-based analysis of the images to both detect and classify suspicious lung lesions; and 4) generation of analysis report. The system was trained using a multi-center dataset of over 2,358 cases of suspicious lung nodules, including cancers and non-cancers. A United States-based, multi-site cohort (n=184) was used for initial algorithm validation where positive cases were defined as biopsy-proven cancer within 1 year, and negatives as negative biopsy or 2+ year negative follow-up. RESULTS: In the validation set, the system achieved a technical completion rate of 100%, automatically for all cases. Median patient age was 62.5 years (interquartile range 58.5 to 66.5 years), 45% were former smokers, 55% were current smokers, and 40% were female. The AI cancer risk (CA-risk) model achieved an area under the receiver-operating-characteristic curve (AUC) of 0.828 [CI: 0.773 – 0.883] in differentiating cancer from non-cancer in cases with known suspicious pulmonary nodules, as confirmed by biopsy, surgery, and/or follow-up. These results were statistically superior to Lung-RADS pAUC of 0.669 [CI: 0.601 – 0.737] and trended better than Brock model AUC of 0.783 [CI: 0.723 – 0.842], Figure 1. At the Youden operating point, model sensitivity was 73.1% [CI: 66.7 – 79.5%], and specificity was 82.5% [CI: 77.0 – 88.0%]. The model's performance was consistent across various CT scanner manufacturers and tested slice thickness. CONCLUSIONS: The closed loop, automated software system, Bronchosolve, consistently processed and analyzed a wide range of CT scans, providing accurate assessment in risk-classifying suspicious lung nodules. This approach has the potential to improve non-invasive assessment of lung nodules eliminating the need for additional user training and time.

Abstract Rationale : Interprofessional providers (IPPs), such as nurses, pharmacists, and respiratory therapists, teaching physician trainees (“interprofessional teaching” [IPT]), may improve teamwork and quality of care in the intensive care unit (ICU). However, the role that IPPs play in physician trainee education in the ICU and their attitudes about IPT have not been well described. In this study, we explored the impact of IPPs on education for Pulmonary and Critical Care fellows and their perceptions regarding IPPs as teachers for medical trainees. Methods: Fellows from eight large academic medical centers with pulmonary and critical care fellowship programs located in the Northeast, Mid-Atlantic, Midwest and West Coast were invited to participate. Eligible participants included Internal Medicine-trained Pulmonary and Critical Care or Critical Care Medicine fellows in the first, second, or third year of fellowship training. All focus groups were led by a single study team member who facilitated sessions in-person at four sites and virtually at four sites. Focus group transcripts were analyzed using a framework approach, and inter-rater reliability testing was performed prior to coding each manuscript in full (Cohen's kappa = 0.77). Results: Focus groups were held at eight sites between March and July 2023. A total of 50 fellows participated, including 13 first year, 21 second year, and 12 third year fellows, all of whom had completed at least 8 weeks of medical ICU rotations during fellowship. Several themes emerged. Fellows identified the relevance of IPP teaching content, fellow bandwidth to learn from IPPs, and the perceived benefit of relationship building with IPPs as factors that made them ideal learners for IPT among medical trainees. Fellows described distinct teaching content from each IPP group, with nurses revealing the “hidden curriculum,” pharmacists providing specific concrete learning points and incorporating primary literature, and respiratory therapists teaching at the bedside. Identified barriers to IPT included fellows’ competing roles as both teacher and learner in limited time, IPPs having difficulty identifying learner needs at varying levels, and fellows’ uncertainty about whether IPPs view themselves as educators. Conclusions: Fellows are uniquely situated to learn from IPPs in the ICU. Perceived benefits include expansion of their knowledge base and fostering relationships with the interprofessional team. The unique teaching profiles of each IPP group could be leveraged to enhance IPT in the ICU.

Abstract Rationale: The 1-year double-blind period of the MANDARA trial (NCT04157348) demonstrated non-inferiority of benralizumab to mepolizumab for achieving remission, in adults with relapsing/refractory eosinophilic granulomatosis with polyangiitis (EGPA). Here, we report the 2-year combined double-blind and ongoing open-label extension (OLE) data. Methods: On completion of the double-blind period (n=136/140) comparing benralizumab 1x30mg versus mepolizumab 3x100mg subcutaneously every 4 weeks, patients were invited to enter the OLE where they continued benralizumab (benra/benra) or switched from mepolizumab to benralizumab (mepo/benra). Endpoints included remission (Birmingham Vasculitis Activity Score=0 and oral glucocorticoid [OGC[ dose ≤4 mg/day), OGC use, relapse, blood eosinophil count (bEOS), Asthma Control Questionnaire (ACQ-6), pre-bronchodilator forced expiratory volume in 1 second (pre-BD FEV1), and safety. Results: In total, 128 patients entered the OLE (n=66 benra/benra, n=62 mepo/benra; mean age 52.8 years; 60.2% female), and 119 completed OLE Year 1. Baseline characteristics of those entering the OLE were similar to the double-blind period. At Week 104, over 60% in both groups (41 [62.1%] benra/benra and 42 [67.7%] mepo/benra patients) achieved remission. During the OLE, 51 (77.3%) benra/benra and 42 (67.7%) mepo/benra patients had no relapses. The percentage of benra/benra patients who discontinued OGCs was similar at Weeks 49-52 (27 [40.9%]) and Weeks 101-104 (29 [43.9%]), while the percentage increased for mepo/benra patients between Weeks 49-52 (16 [25.8%]) and Weeks 101-104 (27 [43.5%]; Figure). The median (IQR) OGC dose at Weeks 101-104 was 0.5 (0,5) in benra/benra and 1.36 (0,5) mg/day in mepo/benra patients. The median (IQR) bEOS in benra/benra patients was 20 (10,40) cells/µL at both Weeks 52 and 100; bEOS were depleted in mepo/benra patients from 70 (40,90) cells/µL to 20 (10,50) cells/µL by 4 weeks after switching. Mean (SD) ACQ-6 scores were 0.64 (0.78) in benra/benra patients and 0.60 (0.76) in mepo/benra patients at Week 104 compared with 1.39 (1.18) and 1.11 (0.95) at baseline. Mean (SD) pre-BD FEV1 was 2.59 (0.90) L versus 2.62 (0.84) L in benra/benra versus mepo/benra patients at Week 52, and 2.61 (0.90) versus 2.63 (0.81) L, at Week 100. Safety of mepolizumab and benralizumab was consistent with their known profiles. Conclusions: In patients with EGPA receiving benralizumab, remission rates, OGC discontinuation, and bEOS depletion were durable over 104 weeks with low relapse rates, and without loss of asthma control or lung function decline. Additional bEOS depletion and OGC sparing was observed in patients switching from mepolizumab to benralizumab.

Abstract Chronic obstructive pulmonary disease (COPD) and pulmonary emphysema were the third-leading cause of death globally in 2019. Besides alpha-1 antitrypsin deficiency, emphysema lacks disease-modifying medications due to its heterogeneity and there remains a need to further investigate emphysema subtypes to advance its diagnosis and treatment. Recent unsupervised machine learning of the texture and anatomical location of emphysema on computed tomography (CT) images of the lung defined six quantitative CT emphysema subtypes uniquely associated with specific gene variants, environmental exposures, symptoms, physiology and clinical events. We aim to identify associations of plasma proteins with each CT emphysema subtype cross-sectionally in a multi-ethnic general population sample. MESA is a prospective cohort study that recruited 6,814 participants from six communities who self-identified as White, Black, Hispanic, or Asian race/ethnicity, ages 45-84 years old and free of clinical cardiovascular disease in 2000-02. The MESA Lung Study performed CT scans among 3,137 participants in 2010-12, which were labeled with previously described (PMID: 37268414) CT emphysema subtypes. Proteomics was assessed using SomaScan version 4.1 assay (7,596 plasma proteins). Linear regression models were adjusted for age, height, weight, sex, race, educational attainment, smoking status, pack years, urine cotinine level, body-mass index (BMI), estimated glomerular filtration rate (eGFR), CT scanner manufacturer and other CT emphysema subtypes. Results were adjusted to reflect a False Discovery Rate (FDR) p-value of less than 5% using the Benjamini-Hochberg method. Among 2,343 participants with both CT and proteomic data (mean age 69.5 years, 52.3% women, 23.7% with COPD), we identified statistically significant associations of proteins with the vanishing lung subtype (42 proteins; top hit Natural cytotoxicity triggering receptor 3 ligand 1, q = 2.9[asterisk]10-9), restrictive combined pulmonary fibrosis/emphysema (CPFE) subtype (11 proteins; top hit Integrin alpha-2/b1, q=1.0[asterisk]10-5), the combined bronchitis-apical emphysema subtype (9 proteins; top hit Desmoglein-2, q=8.8[asterisk]10-13), and the diffuse emphysema subtype (4 proteins; top hit Pulmonary surfactant-associated protein D, q=5.7[asterisk]10-3). No significant proteins were associated with the obstructive CPFE or senile subtypes after FDR correction. Most proteins were uniquely associated with one CT emphysema subtype, while a few proteins were associated with more than one subtype. This study identified specific proteins for different CT emphysema subtypes, many of which have been related less specifically to emphysema, including pulmonary surfactant production, anti-inflammatory pathways, and mitochondrial deregulation, amongst others. Validation of these proteomic associations can provide insight into further pathway assessment, clinical biomarker discovery, and targeted prevention of emphysema in the general population.

Abstract RATIONALE: Airway-to-lung ratio (AirLR), calculated as the mean of airway lumen diameters at standard anatomic locations divided by the cube-root of total lung volume, is associated with COPD risk regardless of smoking status, and we have previously shown an association between total-pulmonary-vascular-volume-to-lung-size relationship with AirLR. [PMCID: PMC11284327] Using dual energy computed tomography (DECT), we recently developed a deep learning-based algorithm for pulmonary arterial segmentation to compute pulmonary-arterial-to-lung ratio (ArtLR) which correlated significantly with AirLR and markers COPD. Here we automate image processing in a larger cohort to estimate ArtLR, introduce an alternative ArtLR standardization and correlate it to COPD-related measures of lung structure and function. METHODS: Using a SPIROMICS sub-cohort (110 participants), imaged via contrast enhanced DECT at functional residual capacity (FRC), an automatic deep learning pipeline segmented the arteries from the pulmonary vascular tree. Another pipeline automatically extracted the centerlines and calculated average arterial segment diameters. A paired non-contrast total lung capacity (TLC) scan was registered to the DECT FRC scan to transfer airway segment labels from the TLC to FRC domain. AirLR was calculated as the mean of airway lumen diameters at standard anatomic locations (trachea-to-subsegments) divided by cube root of lung volume [TLC or FRC]. A simple Long Short-Term Memory (LSTM) network was used to find corresponding anatomic arterial segments. Associations between ArtLR and %Emphysema, AirLR, pre- and post-bronchodilator FEV1/FVC, and a texture-based (AMFM) measure of %broncho-vascular bundles was assessed using linear regression to adjust for age, sex and BMI. RESULTS: In adjusted analyses, a 1-SD decrement in ArtLR-TLC was associated with 4.198 units increase in %Emphysema-910 (95%CI: 1.360-7.037; p = 0.0041), and 0.8198 units decrease in %Bronchovascular (95%CI: 0.5037-1.136; p < 0.0001). A 1-SD increment in ArtLR-TLC was associated with 0.2515 SD increase in AirLR-Outer (95%CI: 0.04947SD-0.4536SD; p = 0.0152), and 0.02083 units increase in PreBronch-FEV1FVC (95%CI: 0.001814-0.03984; p = 0.0321). Sex and BMI show significant associations with ArtLR-TLC, where female is associated with 1.068 SD decrease in ArtLR, and a unit increase in BMI is associated with 0.05788 SD increase in ArtLR (both p-values < 0.0001), and age was not association with ArtLR. CONCLUSIONS: An automated pipeline for pulmonary arterial tree segmentation and standardization from contrast enhanced DECT FRC scans demonstrates a significant relationship between ArtLR and AirLR as well as between ArtLR and %emphysema, and FEV1/FVC. A standardized method of arterial tree size assessment provides for expanded exploration of vascular vs airway disease etiologies.

Abstract Rationale Chronic obstructive pulmonary disease (COPD) manifests later in life and is a leading cause of death worldwide. While smoking-related lung pathology is a long-recognized COPD etiotype, emerging evidence suggests that early-life growth conditions also play an important role. A recently-developed index of early-life growth conditions – the difference between attained height and genotype-predicted height (height-GaP) – has been validated against prospectevely ascertained adverse early-life growth conditions. This study used height-GaP to evaluate the association of early-life growth conditions with COPD phenotype and prognosis later in life. Methods Among current and former smokers with spirometry-defined COPD in the SubPopulations and Intermediate Outcomes Measures in COPD Study (SPIROMICS), we computed height-GaP as the difference between attained height and genotype-predicted height. Regression models were fit for respiratory mortality (defined by primary diagnostic code), dyspnea prevalence (Medical Research Council dyspnea rating >1), chronic bronchitis prevalence, airflow obstruction severity as quantified by FEV1Q, and functional capacity (six-minute walk distance), adjusting for age, age2, sex, race-ethnicity, genotype-predicted height, cigarette smoking status and pack-years. Sensitivity analyses accounted for competing risks of death (Fine-Gray subdistribution hazard model) and accounted for later-life height loss (restricting sample to participants under 50 years). Results Among 1,258 participants with COPD (mean±SD age: 65.4±8.0 years, 42.4% female, mean±SD pack-years: 53.6±28.2, mean±SD FEV1%: 63.3±23.9%), there were 202 respiratory deaths over a median 7.5 years of follow-up (Figure). Larger height-GaP deficit was associated with higher respiratory mortality (adjusted hazard ratio [aHR]: 1.30 [95%CI: 1.12 to 1.50] per 1-SD increment in height-GaP deficit). This association remained significant after accounting for competing events (p=0.001) and when restricted to participants under 50 years (p=0.032). Larger height-GaP deficit was associated with higher dyspnea prevalence (adjusted prevalence ratio: 1.13 [95%CI: 1.02 to 1.25] per 1-SD increment in height-GaP deficit), lower FEV1Q (adjusted mean difference: -0.37 [95%CI: -0.45 to -0.29] per 1-SD height-GaP deficit), and shorter 6-minute walk (adjusted mean difference: -16.6 m [95%CI: -23.1 m to -10.0 m] per 1-SD increment in height-GaP deficit), but not chronic bronchitis or CAT score ≥10 (p>0.05 for each). Conclusions Among current and former smokers with COPD, a quantitative index of early-life conditions was associated with higher respiratory mortality, in addition to dyspnea, airflow obstruction severity and reduced functional capacity. These results suggest that early-life growth conditions may influence COPD phenotype and prognosis later in life.

Abstract Rationale: Application of single cell RNAseq (scRNAseq) in idiopathic pulmonary fibrosis (IPF) and multiple murine models has revealed two unique disease associated fibroblast subsets, the Cthrc1+ fibrotic population, and a population expressing inflammatory markers whose function is unclear. We, and others, have demonstrated that inflammatory fibroblasts are derived from homeostatic fibroblasts and can differentiate to Cthrc1+ fibroblasts. There is also data in mice suggesting that TGFβ inhibition prevents the emergence of Cthrc1+ fibrotic fibroblasts without impacting the inflammatory fibroblast population. Using our inducible genetic murine model of PF, reductionist organoid systems, and scRNAseq from IPF we present evidence and mechanism for inflammatory fibroblasts as a potential regenerative population in the alveolar niche. Methods: Tamoxifen administration to SftpcI73T conditional allelic knock-in mice, induces expression of mutant protein and promotes an inflammatory cascade followed by fibrogenesis. ScRNA-seq and population RNA-seq of bleomycin and SftpcI73T murine models were compared with publicly available data sets from IPF to identify population specific surface marker expression and potential epithelial ligand-receptor pairs. Flow cytometry approaches developed to isolate fibroblast populations were applied for in-vitro organoid modeling and validation of signaling pathways. In-vivo murine intervention in TGFβ signaling was further applied to confirm functional biology of inflammatory fibroblasts. Results: Informatic analysis of human and mouse inflammatory fibroblasts identifies expression of multiple BMP antagonists that have previously been associated with epithelial proliferation. Furthermore, human inflammatory fibroblasts also express key epithelial growth factors including FGF2 and FGF7. Murine organoid modeling confirms that inflammatory fibroblasts support alveolar organoid formation through the inhibition of BMP signaling. As was previously demonstrated in bleomycin, the inhibition of TGFβ signaling in SftpcI73T does not alter the inflammatory fibroblast population, rather this population is sustained, and alveolar epithelial proliferation increases. Additionally, we did not find any evidence for increase inflammation in the mouse lung due to inflammatory fibroblasts preservation. Finally, human data suggests that two populations of inflammatory fibroblasts are present in the lung with only one subset found to express the BMP antagonists and growth factors required to promote alveolar epithelial proliferation. Conclusions: Our data identifies that inflammatory fibroblasts in the alveolar niche inhibit BMP signaling and allow for epithelial expansion. We also determine differences between murine and human inflammatory fibroblasts suggesting that human inflammatory fibroblasts have a greater potential to drive epithelial expansion. These data shed light on the functional role of inflammatory fibroblasts, further refining our understanding of fibroblast heterogeneity in IPF.

Abstract Introduction Medullary thyroid cancer (MTC) is an uncommon malignancy arising from the C-cells of the thyroid gland. It makes up 2% of all thyroid cancers, with both familial and sporadic patterns noted. It is most commonly diagnosed in the fourth to sixth decade of life, with the lungs being a common site of metastasis. We present a case of metastatic MTC with a rare pulmonary presentation. Case Description The patient presented at the age of 28 after having a chronic complaint of stomach pain and diarrhea. Imaging of the abdomen revealed small nodules in the lower pulmonary fields. Dedicated imaging of the chest was subsequently ordered. This study was notable for an enlarged left lobe of the thyroid extending into the mediastinum with two large mediastinal lymph nodes. A fine-needle aspiration of the left neck lymph node was consistent with MTC and there were hematologic findings of elevated calcitonin and CEA. Testing for the RET proto-oncogene was negative. Pathology on total thyroidectomy confirmed the diagnosis, with further sampling identifying metastasis in the thoracic spine (T12). Initial therapy with pazopanib was commenced. After some time on stable therapy, the patient was noted to have waxing and waning pulmonary micro-nodularity associated with increase in his calcitonin levels. There was a finding of a miliary pattern of disease on chest imaging. The pulmonary service was consulted due to these findings in conjunction with sub-acute cough. Pulmonary biopsy was performed, with findings of nodular amyloidosis; Congo red stain highlighted amyloid deposition. Immunohistochemical studies were performed, revealing rare cell groups that were positive for calcitonin, CEA, CAM5.2, TTF-1, chromogranin, and synaptophysin. After further discussion with the Pathology service, the patient was given formal diagnosis of amyloid deposition of the Acal subtype due to metastatic medullary thyroid cancer. Discussion This case highlights a patient diagnosed with sporadic MTC, later found to have metastatic related pulmonary amyloid deposition. While local amyloid deposition can be seen in MTC, it is unusual to have diffuse parenchymal disease due todiffuse parenchymal disease due to distant deposition. There are only four other instances in the literature of a similar presentation. Amyloid deposition is associated with worse outcomes, but the patient had a good outcome with treatment by a tyrosine-kinase inhibitor.

Abstract Introduction: Lung cancer is the most frequently diagnosed cancer and leading cause of cancer death worldwide. While adenocarcinoma is the most common subtype of lung cancer, invasive mucinous adenocarcinoma (IMA) only represents an estimated 2-5% of lung adenocarcinomas. IMA has an incidence of just 0.2% among primary lung cancer patients. This rare form of lung cancer provides a significant diagnostic challenge due to its varied clinical and radiological manifestations, often exhibiting features overlapping with pneumonia. We present a case of IMA with radiographic findings of numerous cavitary nodules, an uncommon morphology reported in 3.5% of cases of IMA, originally thought to represent atypical multifocal pneumonia. Description of Case: A 76-year-old man with a history of coronary artery disease status post percutaneous coronary intervention on dual antiplatelet therapy; hypertension; chronic kidney disease; and 40 pack-year smoking history presented to primary care with 6 weeks of worsening cough productive of clear sputum and shortness of breath. He denied a history of fevers, night sweats, unintentional weight loss, recent travel, or sick contacts. A chest x-ray revealed extensive nodular parenchymal opacities in the lower lobes bilaterally concerning for multifocal pneumonia. Subsequent CT chest showed numerous bilateral, poorly-defined nodules with irregular margins and cavitary lesions. He was admitted for acute hypoxic respiratory failure in the setting of presumed atypical multifocal pneumonia. He completed a course of vancomycin and piperacillin/tazobactam. Notably, bronchoscopy with bronchoalveolar lavage and cytology were negative for infection and malignancy, respectively. Additional noninvasive infectious and inflammatory workup was negative and he was ultimately discharged. He was then readmitted to the ICU for respiratory failure where he eventually underwent transbronchial biopsy. Tissue biopsy revealed moderately-differentiated adenocarcinoma with mucinous features. PD-L1 staining was 0%. Genetic testing revealed two separate TP53 mutations, a KRAS G12D mutation, and SMAD4 alteration. Chemotherapy with carboplatin and pemetrexed was initiated inpatient. He had objective and subjective improvement and was ultimately discharged to continue treatment in the outpatient setting. Discussion: This case represents an uncommon presentation of an already rare form of lung cancer underscoring the diagnostic difficulty in the setting of overlapping clinical and radiological manifestations. Clinicians must consider malignancy when a patient with presumed pneumonia is not improving despite antibiotics. Bronchoscopy with both BAL and transbronchial biopsy is often required to make the diagnosis and exclude other etiologies. Early consideration of alternative diagnoses can improve patient outcomes and help avoid progressive respiratory failure.

Abstract Tracheal tumors are uncommon, comprising only 0.2% of respiratory tract malignancies, with the majority of these being squamous cell carcinoma and adenoid cystic carcinoma. Tracheal chondrosarcomas are a rare cause, with only thirty-four cases described in literature between 1959 and 2020. Although often asymptomatic, nonspecific symptoms including dyspnea, nonproductive cough, and wheezing contribute towards misdiagnosis with diseases such as asthma or chronic obstructive pulmonary disease (COPD). Delay in diagnosis can potentially lead to complete tracheal obstruction and airway compromise. Treatment of choice is surgical resection, with overall 10-year survival to be around 95%. Although recurrence rates are minimal when treated with open surgical resection with negative margins, post-surgical surveillance is recommended. Use of extracorporeal membrane oxygenation (ECMO) during surgery offers an alternative to endotracheal intubation, potentially making complete tumor resection easier. A 56-year-old male never smoker with past medical history of hereditary hemochromatosis, requiring phlebotomy, severe obstructive sleep apnea status post right upper airway nerve stimulator placement, and a history of left vocal cord granuloma that presented to an outpatient pulmonology clinic for chronic cough and exertional dyspnea. Initially, the patient was started on a proton pump inhibitor and nasal corticosteroid, without improvement in symptoms. He subsequently underwent pulmonary function testing, which showed flow volume loop suspicious for variable extrathoracic airway obstruction. Chest computed tomography (CT) scan was significant for a 2.2 cm soft tissue lesion in the left upper mediastinum with protrusion through the tracheal wall. PET scan did not show significant FDG avidity. A 1.5 centimeters lesion situated about 5 centimeters above the carina was found on rigid bronchoscopy and biopsies were negative for malignancy. Patient subsequently underwent video-assisted thoracoscopic surgery (VATS) robotic-assisted tracheal resection on Veno-venous ECMO. Final pathology was consistent with G1 chondrosarcoma, with negative margins. On follow up, patient's respiratory symptoms have resolved and periodic surveillance with repeat chest CTs is planned. This unique case offers an important reminder of a potential cause of upper airway obstruction, not only highlighting the potential of pulmonary function testing to aid in early detection, but also the successful use of ECMO in surgical resection.