Deep learning to detect left ventricular structural abnormalities in chest X-rays.

BACKGROUND AND AIMS: Early identification of cardiac structural abnormalities indicative of heart failure is crucial to improving patient outcomes. Chest X-rays (CXRs) are routinely conducted on a broad population of patients, presenting an opportunity to build scalable screening tools for structural abnormalities indicative of Stage B or worse heart failure with deep learning methods. In this study, a model was developed to identify severe left ventricular hypertrophy (SLVH) and dilated left ventricle (DLV) using CXRs. METHODS: A total of 71 589 unique CXRs from 24 689 different patients completed within 1 year of echocardiograms were identified. Labels for SLVH, DLV, and a composite label indicating the presence of either were extracted from echocardiograms. A deep learning model was developed and evaluated using area under the receiver operating characteristic curve (AUROC). Performance was additionally validated on 8003 CXRs from an external site and compared against visual assessment by 15 board-certified radiologists. RESULTS: The model yielded an AUROC of 0.79 (0.76-0.81) for SLVH, 0.80 (0.77-0.84) for DLV, and 0.80 (0.78-0.83) for the composite label, with similar performance on an external data set. The model outperformed all 15 individual radiologists for predicting the composite label and achieved a sensitivity of 71% vs. 66% against the consensus vote across all radiologists at a fixed specificity of 73%. CONCLUSIONS: Deep learning analysis of CXRs can accurately detect the presence of certain structural abnormalities and may be useful in early identification of patients with LV hypertrophy and dilation. As a resource to promote further innovation, 71 589 CXRs with adjoining echocardiographic labels have been made publicly available.

Advances in CAR T Cell Therapy for Non-Small Cell Lung Cancer.

Since its first approval by the FDA in 2017, tremendous progress has been made in chimeric antigen receptor (CAR) T cell therapy, the adoptive transfer of engineered, CAR-expressing T lymphocyte. CAR T cells are all composed of three main elements: an extracellular antigen-binding domain, an intracellular signaling domain responsible for T cell activation, and a hinge that joins these two domains. Continuous improvement has been made in CARs, now in their fifth generation, particularly in the intracellular signaling domain responsible for T cell activation. CAR T cell therapy has revolutionized the treatment of hematologic malignancies. Nonetheless, the use of CAR T cell therapy for solid tumors has not attained comparable levels of success. Here we review the challenges in achieving effective CAR T cell therapy in solid tumors, and emerging CAR T cells that have shown great promise for non-small cell lung cancer (NSCLC). A growing number of clinical trials have been conducted to study the effect of CAR T cell therapy on NSCLC, targeting different types of surface antigens. They include epidermal growth factor receptor (EGFR), mesothelin (MSLN), prostate stem cell antigen (PSCA), and mucin 1 (MUC1). Potential new targets such as erythropoietin-producing hepatocellular carcinoma A2 (EphA2), tissue factor (TF), and protein tyrosine kinase 7 (PTK7) are currently under investigation in clinical trials. The challenges in developing CAR T for NSCLC therapy and other approaches for enhancing CAR T efficacy are discussed. Finally, we provide our perspective on imaging CAR T cell action by reviewing the two main radionuclide-based CAR T cell imaging techniques, the direct labeling of CAR T cells or indirect labeling via a reporter gene.

Establishing quantitative radiographic criteria for the diagnosis of pleuroparenchymal fibroelastosis.

PURPOSE: Pleuroparenchymal Fibroelastosis (PPFE) is a type of pulmonary fibrosis most commonly occurring at the apices. Patients with PPFE have an increased risk of adverse effects from lung biopsy and in the post-surgical setting. Here, we investigated simple and reproducible measurements on chest CT to evaluate their predictive value in diagnosing PPFE. METHODS: We analyzed a cohort of patients with histologically-proven PPFE and compared them to a cohort of patients diagnosed with "biapical scarring" (BAS) on chest CT. We measured plueuroparenchymal thickness using several independent parameters on chest CT. We also assessed other radiologic and clinical characteristics to identify if any were predictive of PPFF. RESULTS: Our analysis demonstrated the average greatest apical thickness with a cut off of 4.5 mm yielded a sensitivity of 94.4% and a specificity of 88.9%, and an area under the curve of 97.2%. Single greatest apical thickness with a cut off of 7.5 mm had a sensitivity of 100% and a specificity of 88.9%, with the area under the curve of 97.8%. Average greatest upper lobe thickness with a cut off of 8.0 mm had a sensitivity of 88.9% and a specificity of 100%, with an area under the curve of 98.2%. Single greatest upper lobe thickness with a cut off of 8.5 yielded both a sensitivity and specificity of 94.4% and an area under the curve of 94.3%. CONCLUSION: Measurements described above are highly sensitive and specific for the diagnosis of PPFE and warrant investigation with a larger cohort of patients.

Radiographic grading system for usual interstitial pneumonia correlates with mortality and may serve as a surrogate endpoint in clinical trials.

PURPOSE: Usual interstitial pneumonia (UIP)/ idiopathic pulmonary fibrosis (IFP) is a relentlessly progressive lung disease with outcomes similar to cancer. We have previously established a radiologic grading system for UIP and demonstrated that it correlates with pulmonary function tests; here we test the hypothesis that it correlates with mortality. Validating a correlation with mortality will demonstrate the utility of this system for monitoring progression over time clinically and in trials of anti-fibrotic agents. METHODS: We searched the radiology record system "Catalyst" to identify cases and reviewed each case to confirm the diagnosis. 94 patients met the inclusion criteria for further assessment. Chest CT grade was determined by consensus of two cardiothoracic radiologists. Data was analyzed to identify the interval between chest CT and death. This interval was correlated with CT grade using Spearman correlation analysis. RESULTS: For all cases, chest CT grade and mortality demonstrated a positive correlation of rs = 0.37732, 2-tailed p = 0.00018. We also employed subgroup analysis; for the subgroup with intervals less than or equal to 100 days, there was a positive correlation, rs = 0.48339, 2-tailed p = 0.03602; for the subgroup with an interval greater than 100 days between imaging and death there was a positive correlation, rs = 0.302, 2-tailed p = 0.00846. CONCLUSION: These data support use of this system for monitoring clinical progression and as a surrogate endpoint for clinical trials. Future work building upon the data presented here will evaluate its utility in clinical trials and develop automated grading.

Imaging the Side Effects of CAR T Cell Therapy: A Primer for the Practicing Radiologist.

Chimeric antigen receptor (CAR) T cell therapy is a revolutionary form of immunotherapy that has proven to be efficacious in the treatment of many hematologic cancers. CARs are modified T lymphocytes that express an artificial receptor specific to a tumor-associated antigen. These engineered cells are then reintroduced to upregulate the host immune responses and eradicate malignant cells. While the use of CAR T cell therapy is rapidly expanding, little is known about how common side effects such as cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity (ICANS) present radiographically. Here we provide a comprehensive review of how side effects present in different organ systems and how they can be optimally imaged. Early and accurate recognition of the radiographic presentation of these side effects is critical to the practicing radiologist and their patients so that these side effects can be promptly identified and treated.

Advances in PET/CT Imaging for Breast Cancer.

One out of eight women will be affected by breast cancer during her lifetime. Imaging plays a key role in breast cancer detection and management, providing physicians with information about tumor location, heterogeneity, and dissemination. In this review, we describe the latest advances in PET/CT imaging of breast cancer, including novel applications of 18F-FDG PET/CT and the development and testing of new agents for primary and metastatic breast tumor imaging and therapy. Ultimately, these radiopharmaceuticals may guide personalized approaches to optimize treatment based on the patient's specific tumor profile, and may become a new standard of care. In addition, they may enhance the assessment of treatment efficacy and lead to improved outcomes for patients with a breast cancer diagnosis.

Fibrotic-Like Pulmonary Radiographic Patterns Are Not Associated With Adverse Outcomes in COVID-19 Chronic Critical Illness.

OBJECTIVES: Pulmonary fibrosis is a feared complication of COVID-19. To characterize the risks and outcomes associated with fibrotic-like radiographic abnormalities in patients with COVID-19-related acute respiratory distress syndrome (ARDS) and chronic critical illness. DESIGN: Single-center prospective cohort study. SETTING: We examined chest CT scans performed between ICU discharge and 30 days after hospital discharge using established methods to quantify nonfibrotic and fibrotic-like patterns. PATIENTS: Adults hospitalized with COVID-19-related ARDS and chronic critical illness (> 21 d of mechanical ventilation, tracheostomy, and survival to ICU discharge) between March 2020 and May 2020. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: We tested associations of fibrotic-like patterns with clinical characteristics and biomarkers, and with time to mechanical ventilator liberation and 6-month survival, controlling for demographics, comorbidities, and COVID-19 therapies. A total of 141 of 616 adults (23%) with COVID-19-related ARDS developed chronic critical illness, and 64 of 141 (46%) had a chest CT a median (interquartile range) 66 days (42-82 d) after intubation. Fifty-five percent had fibrotic-like patterns characterized by reticulations and/or traction bronchiectasis. In adjusted analyses, interleukin-6 level on the day of intubation was associated with fibrotic-like patterns (odds ratio, 4.40 per quartile change; 95% CI, 1.90-10.1 per quartile change). Other inflammatory biomarkers, Sequential Organ Failure Assessment score, age, tidal volume, driving pressure, and ventilator days were not. Fibrotic-like patterns were not associated with longer time to mechanical ventilator liberation or worse 6-month survival. CONCLUSIONS: Approximately half of adults with COVID-19-associated chronic critical illness have fibrotic-like patterns that are associated with higher interleukin-6 levels at intubation. Fibrotic-like patterns are not associated with longer time to liberation from mechanical ventilation or worse 6-month survival.

Novel Targets, Novel Treatments: The Changing Landscape of Non-Small Cell Lung Cancer.

Treatment of non-small cell lung cancer (NSCLC) has undergone a paradigm shift. Once a disease with limited potential therapies, treatment options for patients have exploded with the availability of molecular testing to direct management and targeted therapies to treat tumors with specific driver mutations. New in vitro diagnostics allow for the early and non-invasive detection of disease, and emerging in vivo imaging techniques allow for better detection and monitoring. The development of checkpoint inhibitor immunotherapy has arguably been the biggest advance in lung cancer treatment, given that the vast majority of NSCLC tumors can be treated with these therapies. Specific targeted therapies, including those against KRAS, EGFR, RTK, and others have also improved the outcomes for those individuals bearing an actionable mutation. New and emerging therapies, such as bispecific antibodies, CAR T cell therapy, and molecular targeted radiotherapy, offer promise to patients for whom none of the existing therapies have proved effective. In this review, we provide the most up-to-date survey to our knowledge regarding emerging diagnostic and therapeutic strategies for lung cancer to provide clinicians with a comprehensive reference of the options for treatment available now and those which are soon to come.

CANNOT B UIP.

The American Thoracic Society provides guidelines for the confident radiographic diagnosis of UIP. In addition, the guidelines identify findings on CT scans that should suggest an alternative diagnosis to UIP. These findings include consolidation, air trapping, nodules, ground glass opacities, cysts, and upper lung and bronchovascular distribution. We present a mnemonic to help the reader remember the list of findings that are inconsistent with UIP and provide imaging examples for completeness. The mnemonic is "CANNOT B" UIP.

More than Just Skin-Deep: A Review of Imaging's Role in Guiding CAR T-Cell Therapy for Advanced Melanoma.

Advanced melanoma is one of the deadliest cancers, owing to its invasiveness and its propensity to develop resistance to therapy. Surgery remains the first-line treatment for early-stage tumors but is often not an option for advanced-stage melanoma. Chemotherapy carries a poor prognosis, and despite advances in targeted therapy, the cancer can develop resistance. CAR T-cell therapy has demonstrated great success against hematological cancers, and clinical trials are deploying it against advanced melanoma. Though melanoma remains a challenging disease to treat, radiology will play an increasing role in monitoring both the CAR T-cells and response to therapy. We review the current imaging techniques for advanced melanoma, as well as novel PET tracers and radiomics, in order to guide CAR T-cell therapy and manage potential adverse events.

The Macklin effect closely correlates with pneumomediastinum in acutely ill intubated patients with COVID-19 infection.

PURPOSE: Patients with COVID-19 infection are frequently found to have pulmonary barotrauma. Recent work has identified the Macklin effect as a radiographic sign that often occurs in patients with COVID-19 and may correlate with barotrauma. METHODS: We evaluated chest CT scans in COVID-19 positive mechanically ventilated patients for the Macklin effect and any type of pulmonary barotrauma. Patient charts were reviewed to identify demographic and clinical characteristics. RESULTS: The Macklin effect on chest CT scan was identified in a total of 10/75 (13.3%) COVID-19 positive mechanically ventilated patients; 9 developed barotrauma. Patients with the Macklin effect on chest CT scan had a 90% rate of pneumomediastinum (p < 0.001) and a trend toward a higher rate of pneumothorax (60%, p = 0.09). Pneumothorax was most frequently omolateral to the site of the Macklin effect (83.3%). CONCLUSION: The Macklin effect may be a strong radiographic biomarker for pulmonary barotrauma, most strongly correlating with pneumomediastinum. Studies in ARDS patients without COVID-19 are needed to validate this sign in a broader population. If validated in a broad population, future critical care treatment algorithms may include the Macklin sign for clinical decision making and prognostication.

Radiographic features of pneumonitis in patients treated with immunotherapy compared to traditional chemotherapy for non-small cell lung cancer.

BACKGROUND: Pneumonitis has been described as a side effect of immunotherapy as well as traditional chemotherapy. Although immune-related adverse event (IRAE) pneumonitis has been extensively characterized, the relationship between IRAE pneumonitis and pneumonitis secondary to chemotherapy is less clear. Here, we present the first analysis of radiographic features of pneumonitis secondary to immunotherapy compared to chemotherapy. METHODS: Using our radiology records system, we searched chest computed tomography (CT) reports for the term "pneumonitis". We evaluated medical records to establish chronicity of pneumonitis occurring after medication administration and excluded cases where radiation therapy appeared to be the cause of pneumonitis. We also obtained information regarding demographic, clinical, and treatment characteristics for comparison. RESULTS: Patients treated with immunotherapy demonstrated more specific features of pneumonitis including consolidation, ground glass opacities, septal thickening, traction bronchiectasis, and pulmonary nodules compared to those treated with chemotherapy. Immunotherapy treatment correlated with the development of pulmonary nodules (p = 0.048), and administration of more than one immunotherapy agent correlated with a greater incidence of development of nodules (p = 0.050). Radiographic features in patients treated with immunotherapy all decreased over time. Conversely, in patients treated with chemotherapy the incidence of ground glass opacities, traction bronchiectasis, pulmonary nodules, and mediastinal/hilar adenopathy increased over time. CONCLUSIONS: IRAE-pneumonitis has distinct features and a distinct clinical course compared to pneumonitis secondary to chemotherapy. Importantly, IRAE-pneumonitis features decreased over time, suggesting that careful consideration of the benefit-risk ratio may allow for continuation of immunotherapy in some patients who develop pneumonitis.

Reporting breast density on chest computed tomography.

Women are encouraged to have a yearly mammogram and in addition to screening for breast cancer, the radiologist reports the patient's breast density. High breast density increases a woman's risk of developing breast cancer. The number of chest computed tomography (CT) scans performed each year is increasing. Chest CT scans for lung cancer screening in high-risk patients are the standard of care. Important additional findings can be identified on these exams including coronary artery calcifications, thyroid nodules, and breast density. Our previous research has shown that breast density can be reliably graded on chest CT and is comparable to mammographic grading. However, the inter-reader agreement was higher for chest CT. It is important that thoracic radiologists include the grading of breast density in their chest CT reports. According to mammography literature, this information has proven to be helpful for early detection of breast cancer. Federal legislation recommends notifying both providers and patients about breast density on mammography and so it follows that if we see the same information on chest CT, we should report it so that at the very least the clinician can encourage their patient to have a routine mammogram.

Chest CT for Breast Cancer Diagnosis.

Background: We report the results of our retrospective analysis of the ability of standard chest CT scans to correctly diagnose cancer in the breast. Methods: Four hundred and fifty-three consecutive women with chest CT scans (contrast and non-contrast) preceding mammograms within one year comprise the study population. All chest CT images were reviewed by an experienced fellowship-trained chest radiologist and mammograms by an experienced fellowship-trained mammographer without the benefit of prior or ancillary studies; only four mammographic views were included for analysis. The size, location, and shape of breast masses were documented; on CT, the average Hounsfield units were measured. On both imaging modalities, the presence of lymphadenopathy, architectural distortion, skin thickening, and microcalcifications were recorded. Ultimately, the interpreting radiologist was asked to decide if a biopsy was indicated, and these recommendations were correlated with the patient's outcome. Findings: Nineteen of four hundred and fifty-three patients had breast cancer at the time of the mammography. Breast masses were the most common finding on chest CT, leading to the recommendation for biopsy. Hounsfield units were the most important feature for discerning benign from malignant masses. CT sensitivity, specificity, and accuracy of CT for breast cancer detection was 84.21%, 99.3%, and 98.68% compared to 78.95%, 93.78%, and 93.16% for four-view mammography. Chest CT scans with or without contrast had similar outcomes for specificity and accuracy, but sensitivity was slightly less without contrast. Chest CT alone, without the benefit of prior exams and patient recall, correctly diagnosed cancer with a p-value of <0.0001 compared to mammography with the same limitations. Conclusion: Chest CT accurately diagnosed breast cancer with few false positives and negatives and did so without the need for patient recall for additional imaging.

Evaluating the Combined Anticancer Response of Checkpoint Inhibitor Immunotherapy and FAP-Targeted Molecular Radiotherapy in Murine Models of Melanoma and Lung Cancer.

Immunotherapy has dramatically improved outcomes for some cancer patients; however, novel treatments are needed for more patients to achieve a long-lasting response. FAP-targeted molecular radiotherapy has shown efficacy in both preclinical and clinical models and has immunomodulatory effects. Here, we studied if combined immunotherapy and radiotherapy could increase antitumor efficacy in murine models of lung cancer and melanoma and interrogated the mechanisms by which these treatments attenuate tumor growth. Using LLC1 and B16F10 murine models of lung cancer and melanoma, respectively, we tested the efficacy of 177Lu-FAPI-04 alone and in combination with immunotherapy. Alone, 177Lu-FAPI-04 significantly reduced tumor growth in both models. In animals with melanoma, combined therapy resulted in tumor regression while lung tumor growth was attenuated, but tumors did not regress. Combined therapy significantly increased caspase-3 and decreased Ki67 compared with immunotherapy alone. Flow cytometry demonstrated that tumor-associated macrophages responded in a tumor-dependent manner which was distinct in animals treated with both therapies compared with either therapy alone. These data demonstrate that 177Lu-FAPI-04 is an effective anticancer therapy for melanoma and lung cancer which mediates effects at least partially through induction of apoptosis and modulation of the immune response. Translational studies with immunotherapy and 177Lu-FAPI-04 are needed to demonstrate the clinical efficacy of this combined regimen.

Impact of the COVID-19 Pandemic on Trends in Cardiothoracic Imaging.

Introduction: Here, we evaluate the effect of the COVID-19 pandemic on utilization of cardiothoracic imaging studies. Methods: We queried our radiology record system to retrospectively identify numbers of specific key cardiothoracic imaging studies for five years prior and during the COVID-19 pandemic. Statistical analysis was performed to evaluate changes in the number of exams in 2020 and 2021 compared to 2019. Results: Five-year retrospective analysis demonstrated progressive increases in nearly all cross-sectional studies. In 2020, daily chest radiograph utilization decreased with an overall number of daily radiographs of 406 (SD = 73.1) compared to 480 per day in 2019 (SD = 82.6) (p < 0.0001). Portable radiograph utilization was increased in 2020 averaging 320 (SD = 68.2) films daily in 2020 compared to 266 (SD = 29.1) in 2019 (p < 0.0001). Utilization of thoracic CT was decreased during the pandemic, with 21.8 (SD = 12.9) studies daily compared to 52.0 (SD = 21.4) (p < 0.0001) studies daily in 2019. Cardiac imaging utilization was also substantially decreased in 2020 compared to 2019, averaging a total of 3.8 (SD = 3.2) versus 10.8 (SD = 6.6) studies daily and 0.88 (SD = 1.7) versus 2.5 (SD = 2.3) studies daily for CT and MRI, respectively. Evaluation of cardiothoracic imaging for the subsequent 18 months after New York's entry to phase I recovery in June 2020 demonstrated that by one year after the emergence of COVID-19 imaging utilization had recovered to prepandemic levels. Cardiac imaging continued to increase throughout the chronic phase of the COVID-19 pandemic, reaching almost twice the prepandemic levels by the end of 2021. Conclusion: COVID-19 has had far-reaching effects on medicine and public health. Here, we demonstrate decreases in all cross-sectional cardiothoracic imaging studies, closely mirroring findings in other fields during the height of the pandemic, which have since rebounded.

COVID-19 on Chest CT: Translating Known Microscopic Findings to Imaging Observations.

PURPOSE: To describe the imaging findings of COVID-19 and correlate them with their known pathology observations. METHODS: This is an IRB-approved retrospective study performed at Columbia University Irving Medical Center (IRB # AAAS9652) that included symptomatic adult patients (21 years or older) who presented to our emergency room and tested positive for COVID-19 and were either admitted or discharged with at least one chest CT from 11 March 2020 through 1 July 2020. CT scans were ordered by the physicians caring for the patients; our COVID-19 care protocols did not specify the timing for chest CT scans. A scoring system was used to document the extent of pulmonary involvement. The total CT grade was the sum of the individual lobar grades and ranged from 0 (no involvement) to 16 (maximum involvement). The distribution of lung abnormalities was described as peripheral (involving the outer one-third of the lung), central (inner two-thirds of the lung), or both. Additional CT findings, including the presence of pleural fluid, atelectasis, fibrosis, cysts, and pneumothorax, were recorded. Contrast-enhanced CT scans were evaluated for the presence of a pulmonary embolism, while non-contrast chest CT scans were evaluated for hyperdense vessels. RESULTS: 209 patients with 232 CT scans met the inclusion criteria. The average age was 61 years (range 23-97 years), and 56% of the patients were male. The average score reflecting the extent of the disease on the CT was 10.2 (out of a potential grade of 16). Further, 73% of the patients received contrast, which allowed the identification of a pulmonary embolism in 21%. Of those without contrast, 33% had hyperdense vessels, which might suggest a chronic pulmonary embolism. Further, 47% had peripheral opacities and 9% had a Hampton's hump, and 78% of the patients had central consolidation, while 28% had round consolidations. Atelectasis was, overall, infrequent at 5%. Fibrosis was observed in 11% of those studied, with 6% having cysts and 3% pneumothorax. CONCLUSIONS: The CT manifestations of COVID-19 can be divided into findings related to endothelial and epithelial injury, as were seen on prior post-mortem reports. Endothelial injury may benefit from treatments to stabilize the endothelium. Epithelial injury is more prone to developing pulmonary fibrotic changes.

Characteristic patterns of SARS-CoV-2 on chest CT suggests a hematologic pathway for viral entry into the lung.

Many SARS-CoV-2 studies have supported the theory that the Type II alveolar epithelial cells (AEC-2) are the primary portal of entry of the virus into the lung following its brief nasal occupation. However, the theory of inhalational transmission of the virus from the ciliated and goblet nasal cells to the lung parenchyma is not supported by the imaging findings on chest computerized tomography (CT), leading the authors to consider an alternative pathway from the nose to the lung parenchyma that could explain the peripheral, basilar predominant pattern of early disease. Imaging supports that the pulmonary capillaries may be an important vehicle for transmission of the virus and/or associated inflammatory mediators to the lung epithelium.