Search

Background It is increasingly recognized that interstitial lung abnormalities (ILAs) detected at CT have potential clinical implications, but automated identification of ILAs has not yet been fully established. Purpose To develop and test automated ILA probability prediction models using machine learning techniques on CT images. Materials and Methods This secondary analysis of a retrospective study included CT scans from patients in the Boston Lung Cancer Study collected between February 2004 and June 2017. Visual assessment of ILAs by two radiologists and a pulmonologist served as the ground truth. Automated ILA probability prediction models were developed that used a stepwise approach involving section inference and case inference models. The section inference model produced an ILA probability for each CT section, and the case inference model integrated these probabilities to generate the case-level ILA probability. For indeterminate sections and cases, both two- and three-label methods were evaluated. For the case inference model, we tested three machine learning classifiers (support vector machine [SVM], random forest [RF], and convolutional neural network [CNN]). Receiver operating characteristic analysis was performed to calculate the area under the receiver operating characteristic curve (AUC). Results A total of 1382 CT scans (mean patient age, 67 years ± 11 [SD]; 759 women) were included. Of the 1382 CT scans, 104 (8%) were assessed as having ILA, 492 (36%) as indeterminate for ILA, and 786 (57%) as without ILA according to ground-truth labeling. The cohort was divided into a training set ( n = 96; ILA, n = 48), a validation set ( n = 24; ILA, n = 12), and a test set ( n = 1262; ILA, n = 44). Among the models evaluated (two- and three-label section inference models; two- and three-label SVM, RF, and CNN case inference models), the model using the three-label method in the section inference model and the two-label method and RF in the case inference model achieved the highest AUC, at 0.87. Conclusion The model demonstrated substantial performance in estimating ILA probability, indicating its potential utility in clinical settings. © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Zagurovskaya in this issue.

Small cell lung cancer (SCLC) is a highly aggressive malignancy with exceptionally poor prognosis, comprising approximately 15% of lung cancers. Emerging knowledge of the molecular and genomic landscape of SCLC and recent successful clinical applications of new systemic agents have allowed for precision oncology treatment approaches. Imaging is essential for the diagnosis, staging, and treatment monitoring of patients with SCLC. The role of imaging is increasing with the approval of new treatment agents, including immune checkpoint inhibitors, which lead to novel imaging manifestations of response and toxicities. The purpose of this state-of-the-art review is to provide the reader with the latest information about SCLC, focusing on the subtyping of this malignancy (molecular characterization) and the emerging systemic therapeutic approaches and their implications for imaging. The review will also discuss the future directions of SCLC imaging, radiomics and machine learning., (© RSNA, 2022.)

Background The clinical impact of interstitial lung abnormalities (ILAs) on poor prognosis has been reported in many studies, but risk stratification in ILA will contribute to clinical practice. Purpose To investigate the association of traction bronchiectasis/bronchiolectasis index (TBI) with mortality and clinical outcomes in individuals with ILA by using the COPDGene cohort. Materials and Methods This study was a secondary analysis of prospectively collected data. Chest CT scans of participants with ILA for traction bronchiectasis/bronchiolectasis were evaluated and outcomes were compared with participants without ILA from the COPDGene study (January 2008 to June 2011). TBI was classified as follows: TBI-0, ILA without traction bronchiectasis/bronchiolectasis; TBI-1, ILA with bronchiolectasis but without bronchiectasis or architectural distortion; TBI-2, ILA with mild to moderate traction bronchiectasis; and TBI-3, ILA with severe traction bronchiectasis and/or honeycombing. Clinical outcomes and overall survival were compared among the TBI groups and the non-ILA group by using multivariable linear regression model and Cox proportional hazards model, respectively. Results Overall, 5295 participants (median age, 59 years; IQR, 52-66 years; 2779 men) were included, and 582 participants with ILA and 4713 participants without ILA were identified. TBI groups were associated with poorer clinical outcomes such as quality of life scores in the multivariable linear regression model (TBI-0: coefficient, 3.2 [95% CI: 0.6, 5.7; P = .01]; TBI-1: coefficient, 3.3 [95% CI: 1.1, 5.6; P = .003]; TBI-2: coefficient, 7.6 [95% CI: 4.0, 11; P < .001]; TBI-3: coefficient, 32 [95% CI: 17, 48; P < .001]). The multivariable Cox model demonstrated that ILA without traction bronchiectasis (TBI-0-1) and with traction bronchiectasis (TBI-2-3) were associated with shorter overall survival (TBI-0-1: hazard ratio [HR], 1.4 [95% CI: 1.0, 1.9; P = .049]; TBI-2-3: HR, 3.8 [95% CI: 2.6, 5.6; P < .001]). Conclusion Traction bronchiectasis/bronchiolectasis was associated with poorer clinical outcomes compared with the group without interstitial lung abnormalities; TBI-2 and 3 were associated with shorter survival. © RSNA, 2022 Online supplemental material is available for this article . See also the editorial by Lee and Im in this issue.

Use of molecular targeting agents and immune checkpoint inhibitors (ICIs) has increased the frequency and broadened the spectrum of lung toxicity, particularly in patients with cancer. The diagnosis of drug-related pneumonitis (DRP) is usually achieved by excluding other potential known causes. Awareness of the incidence and risk factors for DRP is becoming increasingly important. The severity of symptoms associated with DRP may range from mild or none to life-threatening with rapid progression to death. Imaging features of DRP should be assessed in consideration of the distribution of lung parenchymal abnormalities (radiologic pattern approach). The CT patterns reflect acute (diffuse alveolar damage) interstitial pneumonia and transient (simple pulmonary eosinophilia) lung abnormality, subacute interstitial disease (organizing pneumonia and hypersensitivity pneumonitis), and chronic interstitial disease (nonspecific interstitial pneumonia). A single drug can be associated with multiple radiologic patterns. Treatment of a patient suspected of having DRP generally consists of drug discontinuation, immunosuppressive therapy, or both, along with supportive measures eventually including supplemental oxygen and intensive care. In this position paper, the authors provide diagnostic criteria and management recommendations for DRP that should be of interest to radiologists, clinicians, clinical trialists, and trial sponsors, among others. This article is a simultaneous joint publication in Radiology and CHEST . The articles are identical except for stylistic changes in keeping with each journal's style. Either version may be used in citing this article. Published under a CC BY 4.0 license. Online supplemental material is available for this article.

BackgroundImmune checkpoint inhibitors (ICIs) have been increasingly used in cancer treatment, and a subset of patients undergo pseudoprogression. Recognizing the incidence of pseudoprogression is critical for clinical practice.PurposeTo evaluate by systematic review and meta-analysis the incidence of pseudoprogression in cancer treatment with ICIs, and compare the incidence according to response criteria, tumor types, and immunotherapeutic agents.Materials and MethodsMedline and Embase were searched to identify relevant studies published before December 31, 2018. Clinical trials, post hoc analysis of clinical trials, and prospective studies on ICI treatment in patients with malignant solid tumors were included. Pooled incidence of pseudoprogression for all included studies, per definition of pseudoprogression, cancer type, and drug type, was obtained by random-effects models with inverse variance weighting model.ResultsSeventeen studies with 3402 patients were analyzed. The pooled incidence of pseudoprogression was 6.0% (95% confidence interval: 5.0%, 7.0%). The definition of pseudoprogression were divided into four categories: progressive disease followed by partial response (PR) or complete response (CR) but not stable disease (SD) with Response Evaluation Criteria in Solid Tumors (RECIST) 1.1 (six studies); progressive disease followed by SD or PR or CR with RECIST 1.1 (five studies); progressive disease followed by SD or PR or CR with RECIST 1.0 (three studies); and progressive disease followed by SD or PR or CR with immune-related response criteria (irRC) (three studies). Incidence of pseudoprogression varied from 4.5% to 8.0% per definition, ranged from 5.0% to 7.0% per cancer type, and was 5.6% with the monotherapy of programmed cell death-1 inhibitor.ConclusionThe overall incidence of pseudoprogression was 6.0% and was less than 10% in subgroup analyses according to the definitions of pseudoprogression, cancer type, and immune checkpoint inhibitor type. Varying definitions across trials and studies indicates the need for uniform criteria of pseudoprogression for solid tumors.© RSNA, 2020 Online supplemental material is available for this article. See also the article by Dodd and MacDermott in this issue.

Advances in characterization of molecular and genomic abnormalities specific to lung cancer have made precision therapy the current standard of care for lung cancer treatment. This article will provide a cutting-edge review of imaging of lung cancer in the current era of precision medicine. The focus of the article includes (a) an update on the recent advances in precision therapy for non-small cell lung cancer and their implications on imaging; (b) molecular and genomic biomarkers and pitfalls of image interpretations for lung cancer precision therapy; and (c) review of the current approaches and future directions of precision imaging for lung cancer, emphasizing emerging observations in longitudinal tumor kinetics, radiomics, and molecular and functional imaging. The article is designed to help radiologists to remain up to date in the rapidly evolving world of lung cancer therapy and serve as key members of multidisciplinary teams caring for these patients., (© RSNA, 2019.)

Cancer immunotherapy using immune-checkpoint inhibitors has emerged as an effective treatment option for a variety of advanced cancers in the past decade. Because of the distinct mechanisms of immunotherapy that activate the host immunity to treat cancers, unconventional immune-related phenomena are encountered in terms of tumor response and progression, as well as drug toxicity. Imaging plays an important role in objectively characterizing immune-related tumor responses and progression and in detecting and monitoring immune-related adverse events. Moreover, emerging data suggest a promise for molecular imaging that can visualize the specific target molecules involved in immune-checkpoint pathways. In this article, the background and current status of cancer immunotherapy are summarized, and the current methods for imaging evaluations of immune-related responses and toxicities are reviewed along with their limitations and pitfalls. Emerging approaches with molecular imaging are also discussed as a future direction to address unmet needs., (© RSNA, 2018.)

Purpose: To study the frequency, timing, and associations of thoracic metastases in advanced (stage III and IV) high-grade serous ovarian cancer (HGSC) to help optimize the use of cross-sectional chest imaging., Materials and Methods: This institutional review board-approved retrospective study with waived informed consent included 186 consecutive patients with pathologically proven advanced HGSC after primary cytoreduction (mean age ± standard deviation, 60 years ± 9.7) who underwent imaging at our tertiary cancer institution from January 2012 to December 2012 with at least 1 year of follow-up, unless there was thoracic metastasis or death. Electronic medical records and all available imaging studies were reviewed to record patient and tumor characteristics, frequency and timing of abdominal and thoracic metastases, and visibility of the first thoracoabdominal metastasis on abdominal images. Patient and tumor characteristics associated with thoracic metastases were studied by using univariate and multivariate Cox proportional analysis., Results: After median follow-up of 57 months (interquartile range [IQR], 38-93), 175 patients (94%) developed metastatic disease; each had abdominal disease, and 76 (41%) had thoracic metastases. The first thoracoabdominal metastasis was visible on abdominal images in all 175 patients. The thoracic metastasis-free interval was longer than the abdominal disease-free interval (median, 85 months [IQR, 28-131] vs 14 months [IQR, 7-27], respectively; P < .0001). Presence of disease on abdominal images (hazard ratio, 2.56; 95% confidence interval: 1.35, 4.76) was the only factor independently associated with thoracic metastases., Conclusion: Thoracic metastases in advanced HGSC rarely occur before abdominal disease, and first thoracoabdominal metastases are invariably visible on abdominal images. Therefore, cross-sectional chest imaging may be deferred until development of abdominal disease, with minimal risk of missing thoracic metastases.

Tumor response assessment has been a foundation for advances in cancer therapy. Recent discoveries of effective targeted therapy for specific genomic abnormalities in lung cancer and their clinical application have brought revolutionary advances in lung cancer therapy and transformed the oncologist's approach to patients with lung cancer. Because imaging is a major method of response assessment in lung cancer both in clinical trials and practice, radiologists must understand the genomic alterations in lung cancer and the rapidly evolving therapeutic approaches to effectively communicate with oncology colleagues and maintain the key role in lung cancer care. This article describes the origin and importance of tumor response assessment, presents the recent genomic discoveries in lung cancer and therapies directed against these genomic changes, and describes how these discoveries affect the radiology community. The authors then summarize the conventional Response Evaluation Criteria in Solid Tumors and World Health Organization guidelines, which continue to be the major determinants of trial endpoints, and describe their limitations particularly in an era of genomic-based therapy. More advanced imaging techniques for lung cancer response assessment are presented, including computed tomography tumor volume and perfusion, dynamic contrast material-enhanced and diffusion-weighted magnetic resonance imaging, and positron emission tomography with fluorine 18 fluorodeoxyglucose and novel tracers. State-of-art knowledge of lung cancer biology, treatment, and imaging will help the radiology community to remain effective contributors to the personalized care of lung cancer patients., (RSNA, 2014)

Purpose: To retrospectively determine the effect of gestational age (GA), imaging plane, section thickness, and inter- and intraobserver variability on fetal lung volume (FLV) measurements obtained with magnetic resonance (MR) imaging in a cohort of fetuses without thoracic abnormalities., Materials and Methods: Institutional review board approval was obtained. Informed consent for this retrospective cohort study was waived, and the conduct of this study was HIPAA compliant. FLV was measured in 30 fetuses (GA, 17-36 weeks) referred for MR imaging for indications other than pulmonary abnormalities. Measurements were made on single-shot fast spin-echo images by tracing free-form regions of interest on individual consecutive sections in the transverse, sagittal, and coronal planes. Measurements were performed twice by two observers independently. Correlations between FLV and GA, imaging plane, and section thickness were assessed, as were intra- and interobserver variability. Time to perform FLV was assessed in a subset of fetuses., Results: Total FLV ranged from 2 to 110 mL. Mixed-effects regression model showed significant quadratic trend in FLV with increasing GA, with comparable strength of correlation (r = 0.89-0.91) in the three imaging planes of measurement. Intraobserver agreement was good in all three planes (r = 0.65-0.83) and was highest in the transverse plane. Interobserver agreement was good in all three planes (r = 0.68-0.76). FLV showed no significant dependence on section thickness (P = .23) or imaging plane (P = .82). Mean time to obtain FLV measurements ranged from 48 seconds at GA of 21 weeks to 77 seconds at GA of 29-30 weeks., Conclusion: GA-based FLV measurements obtained with MR images are independent of section thickness and imaging plane and can be performed with good inter- and intraobserver agreement in less than 2 minutes., (RSNA, 2006)

Purpose: To compare dynamic expiratory and end-expiratory computed tomography (CT) for depicting central airway collapse in patients with acquired tracheobronchomalacia (TBM)., Materials and Methods: Institutional review board approval was obtained, and informed consent was not needed. Retrospective review was performed of all patients with a CT diagnosis of TBM in a 10-month period (n = 34) who underwent evaluation of airway disease by means of three different sequences at multi-detector row CT: end inspiration, dynamic expiration, and end expiration (the latter was performed only at the levels of the aortic arch, carina, and bronchus intermedius). Fourteen patients (11 men, three women; age range, 19-79 years) who had comparable images obtained with all three sequences at any of these three levels were included in the study. The degree of airway collapse was measured by two thoracic radiologists in consensus by calculating the percentage change in the area of the airway between inspiratory and expiratory scanning. Statistical analysis was performed by using the paired t test., Results: Dynamic expiratory CT elicited a significantly greater degree of airway collapse than end-expiratory CT at all three levels (P < .005). The mean percentages of airway collapse at each of the three levels were as follows: aortic arch, 53.9% with dynamic expiration versus 35.7% with end expiration (P = .0046); carina, 53.6% with dynamic expiration versus 30.9% with end expiration (P < .0001); and bronchus intermedius, 57.5% with dynamic expiration versus 28.6% with end expiration (P = .0022)., Conclusion: Dynamic expiratory CT elicits a significantly greater degree of airway collapse than standard end-expiratory CT in patients with TBM., ((c) RSNA, 2005)