Mediastinal staging for non-small cell lung cancer

Clinical TNM Lung Cancer Staging: A Diagnostic Algorithm with a Pictorial Review

Lung cancer is a prevalent malignant disease with the highest mortality rate among oncological conditions. The assessment of its clinical TNM staging primarily relies on contrast-enhanced computed tomography (CT) of the thorax and proximal abdomen, sometimes with the addition of positron emission tomography/CT scans, mainly for better evaluation of mediastinal lymph node involvement and detection of distant metastases. The purpose of TNM staging is to establish a universal nomenclature for the anatomical extent of lung cancer, facilitating interdisciplinary communication for treatment decisions and research advancements. Recent studies utilizing a large international database and multidisciplinary insights indicate a need to update the TNM classification to enhance the anatomical categorization of lung cancer, ultimately optimizing treatment strategies. The eighth edition of the TNM classification, issued by the International Association for the Study of Lung Cancer (IASLC), transitioned to the ninth edition on 1 January 2025. Key changes include a more detailed classification of the N and M descriptor categories, whereas the T descriptor remains unchanged. Notably, the N2 category will be split into N2a and N2b based on the single-station or multi-station involvement of ipsilateral mediastinal and/or subcarinal lymph nodes, respectively. The M1c category will differentiate between single (M1c1) and multiple (M1c2) organ system involvement for extrathoracic metastases. This review article emphasizes the role of radiologists in implementing the updated TNM classification through CT imaging for correct clinical lung cancer staging and optimal patient management.

Low false-positive lymph nodes for 18 F-fibroblast activation protein inhibitors PET/computed tomography in preoperative staging of patients with nonsmall cell lung cancer

OBJECTIVE

This study aimed to evaluate the diagnostic accuracy of 18 F-fibroblast activation protein inhibitor (FAPI) PET/computed tomography (CT) in identifying primary tumors and mediastinal lymph node metastases in nonsmall cell lung cancer (NSCLC), with histopathological findings serving as the reference standard.

METHODS

Nineteen patients underwent preoperative 18 F-FAPI PET/CT and subsequent surgery; of these, 13 also underwent 18 F-fluorodeoxyglucose (FDG) PET/CT within 1 week. The diagnostic accuracy of primary tumors and lymph node metastases was evaluated for both modalities. Semiquantitative parameters, including maximum standardized uptake values (SUV max ) and target-to-background ratios (TBRs), for both primary tumors and lymph node metastases were assessed for both modalities.

RESULTS

For primary tumors, 18 of 19 (94.7%) showed positive results on 18 F-FAPI PET/CT scans. In 13 patients who also underwent 18 F-FDG PET/CT, 18 F-FAPI PET/CT demonstrated a higher detection rate compared with 18 F-FDG PET/CT (100% vs. 69.1%). The overall accuracy of lymph node assessment with 18 F-FAPI PET/CT (95.9-97.1%) was significantly higher compared to 18 F-FDG PET/CT (51.0%). Malignant lymph nodes exhibited significantly higher SUV max and TBR on 18 F-FAPI scans (SUV max : 7.0 vs. 0.9, P  < 0.001; TBR muscle : 5.0 vs. 0.8, P  < 0.001) than on 18 F-FDG scans (SUV max : 3.9 vs. 1.8, P  = 0.01), except for the liver TBR on 18 F-FDG scans (TBR liver : 1.8 vs. 1.0, P  = 0.055).

CONCLUSION

18 F-FAPI could be utilized in the preoperative staging of NSCLC to mitigate the incidence of false positives associated with 18 F-FDG, due to its higher accuracy in identifying mediastinal lymph node metastasis.

Diagnostic modalities in the mediastinum and the role of bronchoscopy in mediastinal assessment: a narrative review

Background and Objective

Diagnosis of pathology in the mediastinum has proven quite challenging, given the wide variability of both benign and malignant diseases that affect a diverse array of structures. This complexity has led to the development of many different non-invasive and invasive diagnostic modalities. Historically, diagnosis of the mediastinum has relied on different imaging modalities such as chest X-ray, computed tomography (CT), magnetic resonance imaging, and positron emission topography. Once a suspicious lesion was identified with one of these techniques, the gold standard for diagnosis was mediastinoscopy for diagnosis and staging of disease. More recently, many minimally invasive techniques such as CT-guided biopsy, endobronchial ultrasound with transbronchial needle aspiration, and endoscopic ultrasound with fine needle aspiration have revolutionized the diagnosis of the mediastinum. This review provides a comprehensive analysis of all the modalities available for diagnosing mediastinal disease with an emphasis on bronchoscopic techniques.

Methods

Literature search was performed via the PubMed database. We included all types of articles and study designs, including original research, meta-analyses, reviews, and abstracts.

Key Content and Findings

Minimally invasive techniques such as endobronchial ultrasound-transbronchial needle aspiration (EBUS-TBNA) and endoscopic ultrasound-fine needle aspiration (EUS-FNA) have demonstrated high diagnostic yield and low complication rate and have made a significant difference in the time to diagnosis and lives of patients. There continues to be innovation in the field of bronchoscopy with the development of new technologies such as confocal laser endomicroscopy, optical coherence tomography, and artificial intelligence.

Conclusions

Bronchoscopy is and will continue to be an integral modality in minimally invasive diagnosis of the mediastinum.

Clinical utility of artificial intelligence-augmented endobronchial ultrasound elastography in lymph node staging for lung cancer

Objective

Endobronchial ultrasound elastography produces a color map of mediastinal lymph nodes, with the color blue (level 60) indicating stiffness. Our pilot study demonstrated that predominantly blue lymph nodes, with a stiffness area ratio greater than 0.496, are likely malignant. This large-scale study aims to validate this stiffness area ratio compared with pathology.

Methods

This is a single-center prospective clinical trial where B-mode ultrasound and endobronchial ultrasound elastography lymph node images were collected from patients undergoing endobronchial ultrasound transbronchial needle aspiration for suspected or diagnosed non-small cell lung cancer. Images were fed to a trained deep neural network algorithm (NeuralSeg), which segmented the lymph nodes, identified the percent of lymph node area above the color blue threshold of level 60, and assigned a malignant label to lymph nodes with a stiffness area ratio above 0.496. Diagnostic statistics and receiver operating characteristic analyses were conducted. NeuralSeg predictions were compared with pathology.

Results

B-mode ultrasound and endobronchial ultrasound elastography lymph node images (n = 210) were collected from 124 enrolled patients. Only lymph nodes with conclusive pathology results (n = 187) were analyzed. NeuralSeg was able to predict 98 of 143 true negatives and 34 of 44 true positives, resulting in an overall accuracy of 70.59% (95% CI, 63.50-77.01), sensitivity of 43.04% (95% CI, 31.94-54.67), specificity of 90.74% (95% CI, 83.63-95.47), positive predictive value of 77.27% (95% CI, 64.13-86.60), negative predictive value of 68.53% (95% CI, 64.05-72.70), and area under the curve of 0.820 (95% CI, 0.758-0.883).

Conclusions

NeuralSeg was able to predict nodal malignancy based on endobronchial ultrasound elastography lymph node images with high area under the receiver operating characteristic curve and specificity. This technology should be refined further by testing its validity and applicability through a larger dataset in a multicenter trial.

A prospective study to compare the diagnostic accuracy of 99mTc-CNDG SPECT/CT and contrast-enhanced CT in staging of non-small cell lung cancer

OBJECTIVE

To explore the value of 99mTc-isonitrile deoxyglucosamine (CNDG) SPECT/CT in the staging and resectability diagnosis of non-small cell lung cancer (NSCLC) compared with contrast-enhanced CT (CECT).

METHODS

This research was approved by the hospital ethics review committee. Sixty-three patients with NSCLC received 99mTc-CNDG SPECT/CT, CECT and initial TNM staging before treatment. Thirty-three patients who underwent radical surgery underwent postoperative pathological TNM staging as the reference standard. Another thirty patients underwent radiochemotherapy; among them, the reference standard of 7 patients of N staging and 5 patients of M staging was based on biopsy pathology, and the diagnosis of the remaining lesions was confirmed by at least one different image or clinical imaging follow-up for more than 3 months. The McNemar test and receiver operating characteristic (ROC) curve analysis were used to compare the diagnostic accuracy of staging and resectability of 99mTc-CNDG SPECT/CT and CECT in NSCLC, respectively.

RESULTS

For all patients and surgical patients, the accuracies of 99mTc-CNDG SPECT/CT in diagnosing the T stage and N stage were higher than those of CECT (all patients: 90.5%, 88.9% vs. 79.4%, 60.3%; surgical patients: 81.8%, 78.8% vs. 60.6%, 51.5%), and the differences were statistically significant (all patients: T stage, P = 0.016; N stage, P = 0.000; surgical patients: T stage, P = 0.016; N stage, P = 0.004). For all patients, the accuracy of 99mTc-CNDG SPECT/CT in diagnosing the M stage was higher than that of CECT (96.8% vs. 90.5%), but the difference was not statistically significant (P = 0.289). ROC curve analysis showed that the accuracy of 99mTc-CNDG SPECT/CT in diagnosing the potential resectability of NSCLC was significantly better than that of CECT (P = 0.046).

CONCLUSION

This preliminary clinical study shows that 99mTc-CNDG SPECT/CT is of great value for accurate clinical staging of NSCLC compared with CECT and can significantly improve the accuracy of resectability diagnosis.

Artificial Intelligence Algorithm Can Predict Lymph Node Malignancy from Endobronchial Ultrasound Transbronchial Needle Aspiration Images for Non-Small Cell Lung Cancer

INTRODUCTION

Endobronchial ultrasound transbronchial needle aspiration (EBUS-TBNA) for lung cancer staging is operator dependent, resulting in high rates of non-diagnostic lymph node (LN) samples. We hypothesized that an artificial intelligence (AI) algorithm can consistently and reliably predict nodal metastases from the ultrasound images of LNs when compared to pathology.

METHODS

In this analysis of prospectively recorded B-mode images of mediastinal LNs during EBUS-TBNA, we used transfer learning to build an end-to-end ensemble of three deep neural networks (ResNet152V2, InceptionV3, and DenseNet201). Model hyperparameters were tuned, and the optimal version(s) of each model was trained using 80% of the images. A learned ensemble (multi-layer perceptron) of the optimal versions was applied to the remaining 20% of the images (Test Set). All predictions were compared to the final pathology from nodal biopsies and/or surgical specimen.

RESULTS

A total of 2,569 LN images from 773 patients were used. The Training Set included 2,048 LNs, of which 70.02% were benign and 29.98% were malignant on pathology. The Testing Set included 521 LNs, of which 70.06% were benign and 29.94% were malignant on pathology. The final ensemble model had an overall accuracy of 80.63% (95% confidence interval [CI]: 76.93-83.97%), 43.23% sensitivity (95% CI: 35.30-51.41%), 96.91% specificity (95% CI: 94.54-98.45%), 85.90% positive predictive value (95% CI: 76.81-91.80%), 79.68% negative predictive value (95% CI: 77.34-81.83%), and AUC of 0.701 (95% CI: 0.646-0.755) for malignancy.

CONCLUSION

There now exists an AI algorithm which can identify nodal metastases based only on ultrasound images with good overall accuracy, specificity, and positive predictive value. Further optimization with larger sample sizes would be beneficial prior to clinical application.

Sublobar Resection of Non-Small-Cell Lung Cancer: Wedge Resection vs. Segmentectomy

Lung cancer is the most common cause of cancer death. The mainstay treatment for non-small-cell lung cancer (NSCLC), particularly in the early stages, is surgical resection. Traditionally, lobectomy has been considered the gold-standard technique. Sublobar resection includes segmentectomy and wedge resection. Compared to lobectomy, these procedures have been viewed as a compromise procedure, reserved for those with poor cardiopulmonary function or who are poor surgical candidates for other reasons. However, with the advances in imaging and surgical techniques, the subject of sublobar resection as a curative treatment is being revisited. Many studies have now shown segmentectomy to be equivalent to lobectomy in patients with small (<2.0 cm), peripheral NSCLC. However, there is a mix of evidence when it comes to wedge resection and its suitability as a curative procedure. At this time, until more data can be found, segmentectomy should be considered before wedge resection for patients with early-stage NSCLC.

A deep learning-based system for mediastinum station localization in linear EUS (with video)

Background and Objectives

EUS is a crucial diagnostic and therapeutic method for many anatomical regions, especially in the evaluation of mediastinal diseases and related pathologies. Rapidly finding the standard stations is the key to achieving efficient and complete mediastinal EUS imaging. However, it requires substantial technical skills and extensive knowledge of mediastinal anatomy. We constructed a system, named EUS-MPS (EUS-mediastinal position system), for real-time mediastinal EUS station recognition.

Methods

The standard scanning of mediastinum EUS was divided into 7 stations. There were 33 010 images in mediastinum EUS examination collected to construct a station classification model. Then, we used 151 videos clips for video validation and used 1212 EUS images from 2 other hospitals for external validation. An independent data set containing 230 EUS images was applied for the man-machine contest. We conducted a crossover study to evaluate the effectiveness of this system in reducing the difficulty of mediastinal ultrasound image interpretation.

Results

For station classification, the model achieved an accuracy of 90.49% in image validation and 83.80% in video validation. At external validation, the models achieved 89.85% accuracy. In the man-machine contest, the model achieved an accuracy of 84.78%, which was comparable to that of expert (83.91%). The accuracy of the trainees' station recognition was significantly improved in the crossover study, with an increase of 13.26% (95% confidence interval, 11.04%-15.48%; P < 0.05).

Conclusions

This deep learning-based system shows great performance in mediastinum station localization, having the potential to play an important role in shortening the learning curve and establishing standard mediastinal scanning in the future.

The impact of staging FDG-PET/CT on treatment for stage III NSCLC - an analysis of population-based data from Ontario, Canada

Purpose

Fluoro-2-deoxyglucose positron-emission tomography (FDG-PET/CT) is now considered a standard investigation for the staging of new cases of stage III NSCLC. However, there is not published level 3 evidence demonstrating the impact of FDG-PET/CT on appropriate therapy in this setting. Using retrospective population-based data, we sought to examine the role and timing that FDG-PET/CT scans play in influencing treatment choice, as well as survival in patients diagnosed with stage III NSCLC.

Materials and methods

A retrospective cohort of patients diagnosed with stage III NSCLC from 2009-2017 in Ontario were identified from the IC/ES (formerly Institute of Clinical Evaluative Sciences) database. FDG-PET/CT utilization over time, trends in mediastinal biopsy technique and usage, the impact of FDG-PET/CT on overall survival (OS), and its influence on use of concurrent chemoradiotherapy (CRT) were explored. The impact of timing of pre-treatment FDG-PET/CT on OS was also analyzed (≤28 days prior to treatment, 29-56 days prior, and >56 days prior).

Results

Between 2007 and 2017, a total of 13 796 people were diagnosed with stage III NSCLC in Ontario. FDG-PET/CT utilization increased over time with 0% of cases in 2007 and 74% in 2017 with pre-treatment FDG-PET/CT scans. The number of patients who received a mediastinal biopsy similarly increased in this timeframe increasing from 41% to 53%. More patients with pre-treatment FDG-PET/CT scans received curative-intent therapy than those who did not: 23% vs 13% for CRT (p<0.001), and 23% vs 10% for surgery (p<0.001). Median OS was longer in those with FDG-PET/CT scans prior to treatment (17 vs 11 months), as was 5-year survival (22% vs 14%, p<0.001), and this held true on both univariate and multivariate analyses. Timing of FDG-PET/CT scan relative to treatment was not associated with differences in OS.

Conclusion

Improvements in OS were seen in this cohort of stage III NSCLC patients who underwent a pre-treatment FDG-PET/CT scan. This can likely be attributed to stage-appropriate therapy due to more complete staging using FDG-PET/CT. This study stresses the importance of complete staging for suspected stage III NSCLC using FDG-PET/CT, and a need for continued advocacy for increased access to FDG-PET/CT scans.

Application of Artificial Intelligence in the Diagnosis, Treatment, and Prognostic Evaluation of Mediastinal Malignant Tumors

Artificial intelligence (AI), also known as machine intelligence, is widely utilized in the medical field, promoting medical advances. Malignant tumors are the critical focus of medical research and improvement of clinical diagnosis and treatment. Mediastinal malignancy is an important tumor that attracts increasing attention today due to the difficulties in treatment. Combined with artificial intelligence, challenges from drug discovery to survival improvement are constantly being overcome. This article reviews the progress of the use of AI in the diagnosis, treatment, and prognostic prospects of mediastinal malignant tumors based on current literature findings.

Development and validation of a deep learning signature for predicting lymph node metastasis in lung adenocarcinoma: comparison with radiomics signature and clinical-semantic model

OBJECTIVE

To develop and validate a deep learning (DL) signature for predicting lymph node (LN) metastasis in patients with lung adenocarcinoma.

METHODS

A total of 612 patients with pathologically-confirmed lung adenocarcinoma were retrospectively enrolled and were randomly divided into training cohort (n = 489) and internal validation cohort (n = 123). Besides, 108 patients were enrolled and constituted an independent test cohort (n = 108). Patients' clinical characteristics and CT semantic features were collected. The radiomics features were derived from contrast-enhanced CT images. The clinical-semantic model and radiomics signature were built to predict LN metastasis. Furthermore, Swin Transformer was adopted to develop a DL signature predictive of LN metastasis. Model performance was evaluated by area under the receiver operating characteristic curve (AUC), sensitivity, specificity, calibration curve, and decision curve analysis. The comparisons of AUC were conducted by the DeLong test.

RESULTS

The proposed DL signature yielded an AUC of 0.948-0.961 across all three cohorts, significantly superior to both clinical-semantic model and radiomics signature (all p < 0.05). The calibration curves show that DL signature predicted probabilities fit well the actual observed probabilities of LN metastasis. DL signature gained a higher net benefit than both clinical-semantic model and radiomics signature. The incorporation of radiomics signature or clinical-semantic risk predictors failed to reveal an incremental value over the DL signature.

CONCLUSIONS

The proposed DL signature based on Swin Transformer achieved a promising performance in predicting LN metastasis and could confer important information in noninvasive mediastinal LN staging and individualized therapeutic options.

KEY POINTS

• Accurate prediction for lymph node metastasis is crucial to formulate individualized therapeutic options for patients with lung adenocarcinoma. • The deep learning signature yielded an AUC of 0.948-0.961 across all three cohorts in predicting lymph node metastasis, superior to both radiomics signature and clinical-semantic model. • The incorporation of radiomics signature or clinical-semantic risk predictors into deep learning signature failed to reveal an incremental value over deep learning signature.

Is Delayed Image of 18F-FDG PET/CT Necessary for Mediastinal Lymph Node Staging in Non-Small Cell Lung Cancer Patients?

PURPOSE

The purpose of this study was to evaluate the diagnostic accuracies of dual-time-point (DTP) 18F-FDG PET/CT for detection of mediastinal lymph node (LN) metastasis in non-small cell lung cancer (NSCLC) patients through a systematic review and meta-analysis.

PATIENTS AND METHODS

The PubMed, Cochrane database, and EMBASE database, from the earliest available date of indexing through October 31, 2021, were searched for studies evaluating diagnostic performance of DTP 18F-FDG PET/CT for detection of metastatic mediastinal LN in NSCLC patients. We determined the sensitivities and specificities across studies, calculated positive and negative likelihood ratios (LR+ and LR-), and constructed summary receiver operating characteristic curves.

RESULTS

Ten studies (758 patients) were included in the current study. In patient-based analysis, early image showed a sensitivity of 0.76 and a specificity of 0.75. Delayed image revealed a sensitivity of 0.84 and a specificity of 0.71. In LN-based analysis, early image showed a sensitivity of 0.80 and a specificity of 0.83. Delayed image revealed a sensitivity of 0.84 and a specificity of 0.87. Retention index or %ΔSUVmax is superior to early or delayed images of DTP 18F-FDG PET/CT for detection of mediastinal LN metastasis.

CONCLUSIONS

Dual-time-point 18F-FDG PET/CT showed a good diagnostic performances for detection of metastatic mediastinal LNs in NSCLC patients. Early and delayed images of DTP 18F-FDG PET/CT revealed similar diagnostic accuracies for LN metastasis. However, retention index or %ΔSUVmax is superior to early or delayed images of DTP 18F-FDG PET/CT for detection of mediastinal LN metastasis in NSCLC patients. Further large multicenter studies would be necessary to substantiate the diagnostic accuracy of DTP 18F-FDG PET/CT for mediastinal LN staging in NSCLC patients.