Diagnostic accuracy of FDG PET/CT in mediastinal lymph nodes from lung cancer

PET/CT imaging for evaluation of multimodal treatment efficacy and toxicity in advanced NSCLC-current state and future directions

PURPOSE

The advent of immune checkpoint inhibitors (ICIs) has revolutionized the treatment of advanced NSCLC, leading to a string of approvals in recent years. Herein, a narrative review on the role of 18F-fluorodeoxyglucose positron emission tomography/computed tomography (FDG PET/CT) in the ever-evolving treatment landscape of advanced NSCLC is presented.

METHODS

This comprehensive review will begin with an introduction into current treatment paradigms incorporating ICIs; the evolution of CT-based criteria; moving onto novel phenomena observed with ICIs and the current state of hybrid imaging for diagnosis, treatment planning, evaluation of treatment efficacy and toxicity in advanced NSCLC, also taking into consideration its limitations and future directions.

CONCLUSIONS

The advent of ICIs marks the dawn of a new era bringing forth new challenges particularly vis-à-vis treatment response assessment and observation of novel phenomena accompanied by novel systemic side effects. While FDG PET/CT is widely adopted for tumor volume delineation in locally advanced disease, response assessment to immunotherapy based on current criteria is of high clinical value but has its inherent limitations. In recent years, modifications of established (PET)/CT criteria have been proposed to provide more refined approaches towards response evaluation. Not only a comprehensive inclusion of PET-based response criteria in prospective randomized controlled trials, but also a general harmonization within the variety of PET-based response criteria is pertinent to strengthen clinical implementation and widespread use of hybrid imaging for response assessment in NSCLC.

Challenges in the target volume definition of lung cancer radiotherapy

Radiotherapy, with or without systemic treatment has an important role in the management of lung cancer. In order to deliver the treatment accurately, the clinician must precisely outline the gross tumour volume (GTV), mostly on computed tomography (CT) images. However, due to the limited contrast between tumour and non-malignant changes in the lung tissue, it can be difficult to distinguish the tumour boundaries on CT images leading to large interobserver variation and differences in interpretation. Therefore the definition of the GTV has often been described as the weakest link in radiotherapy with its inaccuracy potentially leading to missing the tumour or unnecessarily irradiating normal tissue. In this article, we review the various techniques that can be used to reduce delineation uncertainties in lung cancer.

Prospective comparison of 18F-FDG PET/MRI and 18F-FDG PET/CT for thoracic staging of non-small cell lung cancer

OBJECTIVES

To compare the diagnostic performance of ¹⁸F-FDG PET/MRI and ¹⁸F-FDG PET/CT for primary and locoregional lymph node staging in non-small cell lung cancer (NSCLC).

METHODS

In this prospective study, a total of 84 patients (51 men, 33 women, mean age 62.5 ± 9.1 years) with histopathologically confirmed NSCLC underwent ¹⁸F-FDG PET/CT followed by ¹⁸F-FDG PET/MRI in a single injection protocol. Two readers independently assessed T and N staging in separate sessions according to the seventh edition of the American Joint Committee on Cancer staging manual for ¹⁸F-FDG PET/CT and ¹⁸F-FDG PET/MRI, respectively. Histopathology as a reference standard was available for N staging in all 84 patients and for T staging in 39 patients. Differences in staging accuracy were assessed by McNemars chi² test. The maximum standardized uptake value (SUV_max) and longitudinal diameters of primary tumors were correlated using Pearson's coefficients.

RESULTS

T stage was categorized concordantly in ¹⁸F-FDG PET/MRI and ¹⁸F-FDG PET/CT in 38 of 39 (97.4%) patients. Herein, ¹⁸F-FDG PET/CT and ¹⁸F-FDG PET/MRI correctly determined the T stage in 92.3 and 89.7% of patients, respectively. N stage was categorized concordantly in 83 of 84 patients (98.8%). ¹⁸F-FDG PET/CT correctly determined the N stage in 78 of 84 patients (92.9%), while ¹⁸F-FDG PET/MRI correctly determined the N stage in 77 of 84 patients (91.7%). Differences between ¹⁸F-FDG PET/CT and ¹⁸F-FDG PET/MRI in T and N staging accuracy were not statistically significant (p > 0.5, each). Tumor size and SUV_max measurements derived from both imaging modalities exhibited excellent correlation (r = 0.963 and r = 0.901, respectively).

CONCLUSION

¹⁸F-FDG PET/MRI and ¹⁸F-FDG PET/CT show an equivalently high diagnostic performance for T and N staging in patients suffering from NSCLC.

Comparison of machine learning methods for classifying mediastinal lymph node metastasis of non-small cell lung cancer from 18F-FDG PET/CT images

BACKGROUND

This study aimed to compare one state-of-the-art deep learning method and four classical machine learning methods for classifying mediastinal lymph node metastasis of non-small cell lung cancer (NSCLC) from 18F-FDG PET/CT images. Another objective was to compare the discriminative power of the recently popular PET/CT texture features with the widely used diagnostic features such as tumor size, CT value, SUV, image contrast, and intensity standard deviation. The four classical machine learning methods included random forests, support vector machines, adaptive boosting, and artificial neural network. The deep learning method was the convolutional neural networks (CNN). The five methods were evaluated using 1397 lymph nodes collected from PET/CT images of 168 patients, with corresponding pathology analysis results as gold standard. The comparison was conducted using 10 times 10-fold cross-validation based on the criterion of sensitivity, specificity, accuracy (ACC), and area under the ROC curve (AUC). For each classical method, different input features were compared to select the optimal feature set. Based on the optimal feature set, the classical methods were compared with CNN, as well as with human doctors from our institute.

RESULTS

For the classical methods, the diagnostic features resulted in 81~85% ACC and 0.87~0.92 AUC, which were significantly higher than the results of texture features. CNN's sensitivity, specificity, ACC, and AUC were 84, 88, 86, and 0.91, respectively. There was no significant difference between the results of CNN and the best classical method. The sensitivity, specificity, and ACC of human doctors were 73, 90, and 82, respectively. All the five machine learning methods had higher sensitivities but lower specificities than human doctors.

CONCLUSIONS

The present study shows that the performance of CNN is not significantly different from the best classical methods and human doctors for classifying mediastinal lymph node metastasis of NSCLC from PET/CT images. Because CNN does not need tumor segmentation or feature calculation, it is more convenient and more objective than the classical methods. However, CNN does not make use of the import diagnostic features, which have been proved more discriminative than the texture features for classifying small-sized lymph nodes. Therefore, incorporating the diagnostic features into CNN is a promising direction for future research.