Accuracy of the Vancouver Lung Cancer Risk Prediction Model Compared With That of Radiologists

Imaging of Solid Pulmonary Nodules

Early detection with accurate classification of solid pulmonary nodules is critical in reducing lung cancer morbidity and mortality. Computed tomography (CT) remains the most widely used imaging examination for pulmonary nodule evaluation; however, other imaging modalities, such as PET/CT and MRI, are increasingly used for nodule characterization. Current advances in solid nodule imaging are largely due to developments in machine learning, including automated nodule segmentation and computer-aided detection. This review explores current multi-modality solid pulmonary nodule detection and characterization with discussion of radiomics and risk prediction models.

Radiomics and artificial intelligence for risk stratification of pulmonary nodules: Ready for primetime?

Pulmonary nodules are ubiquitously found on computed tomography (CT) imaging either incidentally or via lung cancer screening and require careful diagnostic evaluation and management to both diagnose malignancy when present and avoid unnecessary biopsy of benign lesions. To engage in this complex decision-making, clinicians must first risk stratify pulmonary nodules to determine what the best course of action should be. Recent developments in imaging technology, computer processing power, and artificial intelligence algorithms have yielded radiomics-based computer-aided diagnosis tools that use CT imaging data including features invisible to the naked human eye to predict pulmonary nodule malignancy risk and are designed to be used as a supplement to routine clinical risk assessment. These tools vary widely in their algorithm construction, internal and external validation populations, intended-use populations, and commercial availability. While several clinical validation studies have been published, robust clinical utility and clinical effectiveness data are not yet currently available. However, there is reason for optimism as ongoing and future studies aim to target this knowledge gap, in the hopes of improving the diagnostic process for patients with pulmonary nodules.

Development of a TLR-Based Model That Can Predict Prognosis, Tumor Microenvironment, and Drug Response for Esophageal Squamous Cell Carcinoma

The toll-like receptor (TLR) family is an important class of proteins involved in the immune response. However, little is known about the association between TLRs and Esophageal squamous cell cancer (ESCC). We explored differentially expressed genes (DEGs) between ESCC and esophagus tissues in TCGA and GTEx database. By taking the intersection with TLR gene set and using univariate Cox analysis and multivariate Cox regression analysis to discriminate the hub genes, we created a TLR-prognostic model. Our model separated patients with ESCC into high- and low-risk score (RS) groups. Prognostic analysis was performed with Kaplan-Meier curves. The two groups were also compared regarding tumor immune microenvironment and drug sensitivity. Six hub genes (including CD36, LGR4, MAP2K3, NINJ1, PIK3R1, and TRAF3) were screened to construct a TLR-prognostic model. High-RS group had a worse survival (p < 0.01), lower immune checkpoint expression (p < 0.05), immune cell abundance (p < 0.05) and decreased sensitivity to Epirubicin (p < 0.001), 5-fluorouracil (p < 0.0001), Sorafenib (p < 0.01) and Oxaliplatin (p < 0.05). We constructed a TLR-based model, which could be used to assess the prognosis of patients with ESCC, provide new insights into drug treatment for ESCC patients and investigate the TME and drug response.

Artificial Intelligence Tool for Assessment of Indeterminate Pulmonary Nodules Detected with CT

Background Limited data are available regarding whether computer-aided diagnosis (CAD) improves assessment of malignancy risk in indeterminate pulmonary nodules (IPNs). Purpose To evaluate the effect of an artificial intelligence-based CAD tool on clinician IPN diagnostic performance and agreement for both malignancy risk categories and management recommendations. Materials and Methods This was a retrospective multireader multicase study performed in June and July 2020 on chest CT studies of IPNs. Readers used only CT imaging data and provided an estimate of malignancy risk and a management recommendation for each case without and with CAD. The effect of CAD on average reader diagnostic performance was assessed using the Obuchowski-Rockette and Dorfman-Berbaum-Metz method to calculate estimates of area under the receiver operating characteristic curve (AUC), sensitivity, and specificity. Multirater Fleiss κ statistics were used to measure interobserver agreement for malignancy risk and management recommendations. Results A total of 300 chest CT scans of IPNs with maximal diameters of 5-30 mm (50.0% malignant) were reviewed by 12 readers (six radiologists, six pulmonologists) (patient median age, 65 years; IQR, 59-71 years; 164 [55%] men). Readers' average AUC improved from 0.82 to 0.89 with CAD (P < .001). At malignancy risk thresholds of 5% and 65%, use of CAD improved average sensitivity from 94.1% to 97.9% (P = .01) and from 52.6% to 63.1% (P < .001), respectively. Average reader specificity improved from 37.4% to 42.3% (P = .03) and from 87.3% to 89.9% (P = .05), respectively. Reader interobserver agreement improved with CAD for both the less than 5% (Fleiss κ, 0.50 vs 0.71; P < .001) and more than 65% (Fleiss κ, 0.54 vs 0.71; P < .001) malignancy risk categories. Overall reader interobserver agreement for management recommendation categories (no action, CT surveillance, diagnostic procedure) also improved with CAD (Fleiss κ, 0.44 vs 0.52; P = .001). Conclusion Use of computer-aided diagnosis improved estimation of indeterminate pulmonary nodule malignancy risk on chest CT scans and improved interobserver agreement for both risk stratification and management recommendations. © RSNA, 2022 Online supplemental material is available for this article. See also the editorial by Yanagawa in this issue.

Evaluating the Patient With a Pulmonary Nodule: A Review

IMPORTANCE

Pulmonary nodules are identified in approximately 1.6 million patients per year in the US and are detected on approximately 30% of computed tomographic (CT) images of the chest. Optimal treatment of an individual with a pulmonary nodule can lead to early detection of cancer while minimizing testing for a benign nodule.

OBSERVATIONS

At least 95% of all pulmonary nodules identified are benign, most often granulomas or intrapulmonary lymph nodes. Smaller nodules are more likely to be benign. Pulmonary nodules are categorized as small solid (<8 mm), larger solid (≥8 mm), and subsolid. Subsolid nodules are divided into ground-glass nodules (no solid component) and part-solid (both ground-glass and solid components). The probability of malignancy is less than 1% for all nodules smaller than 6 mm and 1% to 2% for nodules 6 mm to 8 mm. Nodules that are 6 mm to 8 mm can be followed with a repeat chest CT in 6 to 12 months, depending on the presence of patient risk factors and imaging characteristics associated with lung malignancy, clinical judgment about the probability of malignancy, and patient preferences. The treatment of an individual with a solid pulmonary nodule 8 mm or larger is based on the estimated probability of malignancy; the presence of patient comorbidities, such as chronic obstructive pulmonary disease and coronary artery disease; and patient preferences. Management options include surveillance imaging, defined as monitoring for nodule growth with chest CT imaging, positron emission tomography-CT imaging, nonsurgical biopsy with bronchoscopy or transthoracic needle biopsy, and surgical resection. Part-solid pulmonary nodules are managed according to the size of the solid component. Larger solid components are associated with a higher risk of malignancy. Ground-glass pulmonary nodules have a probability of malignancy of 10% to 50% when they persist beyond 3 months and are larger than 10 mm in diameter. A malignant nodule that is entirely ground glass in appearance is typically slow growing. Current bronchoscopy and transthoracic needle biopsy methods yield a sensitivity of 70% to 90% for a diagnosis of lung cancer.

CONCLUSIONS AND RELEVANCE

Pulmonary nodules are identified in approximately 1.6 million people per year in the US and approximately 30% of chest CT images. The treatment of an individual with a pulmonary nodule should be guided by the probability that the nodule is malignant, safety of testing, the likelihood that additional testing will be informative, and patient preferences.

Surgery without preoperative histological confirmation of lung cancer: what is the current clinical practice?

Background

There are discordances in the guidelines regarding the need to acquire histological diagnosis before surgical treatment of (presumed) lung cancer. Preoperative histological confirmation is always encouraged in this setting to prevent unnecessary surgery or when sublobar resection for small-sized tumors is considered. The aim of this retrospective cohort study was to assess the proportion of patients undergoing lung cancer resection in the Netherlands without preoperative pathological confirmation, based on the intraoperative pathological diagnosis (IOD) rate, and to determine characteristics that may influence IOD frequency.

Methods

Data on 10,226 patients, who underwent surgical treatment for lung cancer from 2010 to 2015, were retrieved from the Netherlands National Cancer Registry. We registered an IOD when the date of diagnosis equaled the date of the first surgical intervention. Tabulations and multivariable logistic regression were used to identify predictive parameters for IOD.

Results

36% of surgical procedures were classified as IOD, and decreased with increasing tumor size and extent of surgery (57% for segmentectomy, 39% for lobectomy and 11% for pneumonectomy). IOD was more frequently observed in adenocarcinoma (41%), varied between hospitals from 13% to 66% and was less common when patients were referred from a hospital where thoracic surgery was not performed. Previous history of cancer did not affect IOD.

Conclusions

More than one-third of patients with suspected lung cancer in the Netherlands was operated without preoperative histological confirmation. There was significant variation in IOD rates between different hospitals, which deserves further detailed analysis when striving for uniform surgical quality of care for patients with lung cancer.

On the performance of lung nodule detection, segmentation and classification

Computed tomography (CT) screening is an effective way for early detection of lung cancer in order to improve the survival rate of such a deadly disease. For more than two decades, image processing techniques such as nodule detection, segmentation, and classification have been extensively studied to assist physicians in identifying nodules from hundreds of CT slices to measure shapes and HU distributions of nodules automatically and to distinguish their malignancy. Thanks to new parallel computation, multi-layer convolution, nonlinear pooling operation, and the big data learning strategy, recent development of deep-learning algorithms has shown great progress in lung nodule screening and computer-assisted diagnosis (CADx) applications due to their high sensitivity and low false positive rates. This paper presents a survey of state-of-the-art deep-learning-based lung nodule screening and analysis techniques focusing on their performance and clinical applications, aiming to help better understand the current performance, the limitation, and the future trends of lung nodule analysis.