Discrimination of mediastinal metastatic lymph nodes in NSCLC based on radiomic features in different phases of CT imaging

Incidence rate of occult lymph node metastasis in clinical T1-2N0M0 small cell lung cancer patients and radiomic prediction based on contrast-enhanced CT imaging: a multicenter study : Original research

BACKGROUND

This study aimed to explore the incidence of occult lymph node metastasis (OLM) in clinical T1 - 2N0M0 (cT1 - 2N0M0) small cell lung cancer (SCLC) patients and develop machine learning prediction models using preoperative intratumoral and peritumoral contrast-enhanced CT-based radiomic data.

METHODS

By conducting a retrospective analysis involving 242 eligible patients from 4 centeres, we determined the incidence of OLM in cT1 - 2N0M0 SCLC patients. For each lesion, two ROIs were defined using the gross tumour volume (GTV) and peritumoral volume 15 mm around the tumour (PTV). By extracting a comprehensive set of 1595 enhanced CT-based radiomic features individually from the GTV and PTV, five models were constucted and we rigorously evaluated the model performance using various metrics, including the area under the curve (AUC), accuracy, sensitivity, specificity, calibration curve, and decision curve analysis (DCA). For enhanced clinical applicability, we formulated a nomogram that integrates clinical parameters and the rad_score (GTV and PTV).

RESULTS

The initial investigation revealed a 33.9% OLM positivity rate in cT1 - 2N0M0 SCLC patients. Our combined model, which incorporates three radiomic features from the GTV and PTV, along with two clinical parameters (smoking status and shape), exhibited robust predictive capabilities. With a peak AUC value of 0.772 in the external validation cohort, the model outperformed the alternative models. The nomogram significantly enhanced diagnostic precision for radiologists and added substantial value to the clinical decision-making process for cT1 - 2N0M0 SCLC patients.

CONCLUSIONS

The incidence of OLM in SCLC patients surpassed that in non-small cell lung cancer patients. The combined model demonstrated a notable generalization effect, effectively distinguishing between positive and negative OLMs in a noninvasive manner, thereby guiding individualized clinical decisions for patients with cT1 - 2N0M0 SCLC.

Prediction Models for Mediastinal Metastasis and Its Detection by Endobronchial Ultrasound-Guided Transbronchial Needle Aspiration in Potentially Operable Non-Small Cell Lung Cancer: A Prospective Study

BACKGROUND

Prediction models for mediastinal metastasis and its detection by endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) have not been developed using a prospective cohort of potentially operable patients with non-small cell lung cancer (NSCLC).

RESEARCH QUESTION

Can mediastinal metastasis and its detection by EBUS-TBNA be predicted with prediction models in NSCLC?

STUDY DESIGN AND METHODS

For the prospective development cohort, 589 potentially operable patients with NSCLC were evaluated (July 2016-June 2019) from five Korean teaching hospitals. Mediastinal staging was performed using EBUS-TBNA (with or without the transesophageal approach). Surgery was performed for patients without clinical N (cN) 2-3 disease by endoscopic staging. The prediction model for lung cancer staging-mediastinal metastasis (PLUS-M) and a model for mediastinal metastasis detection by EBUS-TBNA (PLUS-E) were developed using multivariable logistic regression analyses. Validation was performed using a retrospective cohort (n = 309) from a different period (June 2019-August 2021).

RESULTS

The prevalence of mediastinal metastasis diagnosed by EBUS-TBNA or surgery and the sensitivity of EBUS-TBNA in the development cohort were 35.3% and 87.0%, respectively. In PLUS-M, younger age (< 60 years and 60-70 years compared with ≥ 70 years), nonsquamous histology (adenocarcinoma and others), central tumor location, tumor size (> 3-5 cm), cN1 or cN2-3 stage by CT, and cN1 or cN2-3 stage by PET-CT were significant risk factors for N2-3 disease. Areas under the receiver operating characteristic curve (AUCs) for PLUS-M and PLUS-E were 0.876 (95% CI, 0.845-0.906) and 0.889 (95% CI, 0.859-0.918), respectively. Model fit was good (PLUS-M: Hosmer-Lemeshow P = .658, Brier score = 0.129; PLUS-E: Hosmer-Lemeshow P = .569, Brier score = 0.118). In the validation cohort, PLUS-M (AUC, 0.859 [95% CI, 0.817-0.902], Hosmer-Lemeshow P = .609, Brier score = 0.144) and PLUS-E (AUC, 0.900 [95% CI, 0.865-0.936], Hosmer-Lemeshow P = .361, Brier score = 0.112) showed good discrimination ability and calibration.

INTERPRETATION

PLUS-M and PLUS-E can be used effectively for decision-making for invasive mediastinal staging in NSCLC.

TRIAL REGISTRY

ClinicalTrials.gov; No.: NCT02991924; URL: www.

CLINICALTRIALS

gov.

Morphometry-based radiomics for predicting therapeutic response in patients with gliomas following radiotherapy

Introduction

Gliomas are still considered as challenging in oncologic management despite the developments in treatment approaches. The complete elimination of a glioma might not be possible even after a treatment and assessment of therapeutic response is important to determine the future course of actions for patients with such cancers. In the recent years radiomics has emerged as a promising solution with potential applications including prediction of therapeutic response. Hence, this study was focused on investigating whether morphometry-based radiomics signature could be used to predict therapeutic response in patients with gliomas following radiotherapy.

Methods

105 magnetic resonance (MR) images including segmented and non-segmented images were used to extract morphometric features and develop a morphometry-based radiomics signature. After determining the appropriate machine learning algorithm, a prediction model was developed to predict the therapeutic response eliminating the highly correlated features as well as without eliminating the highly correlated features. Then the model performance was evaluated.

Results

Tumor grade had the highest contribution to develop the morphometry-based signature. Random forest provided the highest accuracy to train the prediction model derived from the morphometry-based radiomics signature. An accuracy of 86% and area under the curve (AUC) value of 0.91 were achieved for the prediction model evaluated without eliminating the highly correlated features whereas accuracy and AUC value were 84% and 0.92 respectively for the prediction model evaluated after eliminating the highly correlated features.

Discussion

Nonetheless, the developed morphometry-based radiomics signature could be utilized as a noninvasive biomarker for therapeutic response in patients with gliomas following radiotherapy.

Preoperative CT findings and prognosis of pulmonary sarcomatoid carcinoma: comparison with conventional NSCLC of similar tumor size

BACKGROUND

Pulmonary sarcomatoid carcinoma (PSC) is a rare subtype of non-small cell lung cancer (NSCLC) but differs in terms of treatment strategies compared with conventional-NSCLC (c-NSCLC). However, preoperative CT differentiation between PSC and c-NSCLC remains a challenge. This study aimed to explore the CT findings and prognosis of PSC compared with c-NSCLC of similar tumor size.

METHODS

Clinical data and CT findings of 31 patients with PSC and 87 patients with c-NSCLC were retrospectively analyzed. Clinical data included sex, age, and smoking history. CT findings included tumor size, tumor location, calcification, vacuole/cavity, pleural invasion, mean CT value, and low-attenuation area (LAA) ratio. Kaplan‒Meier curves and log-rank tests were used for survival analysis. A Cox regression model was constructed to evaluate prognostic risk factors associated with overall survival (OS). The Spearman correlation among clinicoradiological outcomes were analyzed.

RESULTS

The mean tumor size of PSC and c-NSCLC were both 5.1 cm. The median survival times of PSC and c-NSCLC were 8 months and 34 months, respectively (P < 0.001). Calcification and vacuoles/cavities were rarely present in PSC. Pleural invasion occurred in both PSC and c-NSCLC (P = 0.285). The mean CT values of PSC and c-NSCLC on plain scan (PS), arterial phase (AP), and venous phase (VP) were 30.48 ± 1.59 vs. 36.25 ± 0.64 Hu (P = 0.002), 43.26 ± 2.96 vs. 58.71 ± 1.65 Hu (P < 0.001) and 50.26 ± 3.28 vs. 64.24 ± 1.86 Hu (P < 0.001), the AUCs were 0.685, 0.757 and 0.710, respectively. Compared to c-NSCLC, PSC had a larger LAA ratio, and the AUC was 0.802, with an optimal cutoff value of 20.6%, and the sensitivity and specificity were 0.645 and 0.862, respectively. Combined with the mean CT value and LAA ratio, AP + VP + LAA yielded the largest AUC of 0.826. The LAA ratio were not independent risk factors for PSC in this study. LAA ratio was negatively correlated with PS (r = -0.29), AP (r = -0.58), and VP (r = -0.66). LAA showed a weak positive correlation with tumor size(r = 0.27).

CONCLUSIONS

PSC has a poorer prognosis than c-NSCLC of similar tumor size. The mean CT value and LAA ratio contributes to preoperative CT differentiation of PSC and c-NSCLC.

[Research Progress in Imaging-based Diagnosis of Benign and Malignant Enlarged Lymph Nodes in Non-small Cell Lung Cancer]

Non-small cell lung cancer (NSCLC) can be detected with enlarged lymph nodes on imaging, but their benignity and malignancy are difficult to determine directly, making it difficult to stage the tumor and design radiotherapy target volumes. The clinical diagnosis of malignant lymph nodes is often based on the short diameter of lymph nodes ≥1 cm or the maximum standard uptake value ≥2.5, but the sensitivity and specificity of these criteria are too low to meet the clinical needs. In recent years, many advances have been made in diagnosing benign and malignant lymph nodes using other imaging parameters, and with the development of radiomics, deep learning and other technologies, models of mining the image information of enlarged lymph node regions further improve the diagnostic accuracy. The purpose of this paper is to review recent advances in imaging-based diagnosis of benign and malignant enlarged lymph nodes in NSCLC for more accurate and noninvasive assessment of lymph node status in clinical practice.
.

Integrating Preoperative Computed Tomography and Clinical Factors for Lymph Node Metastasis Prediction in Esophageal Squamous Cell Carcinoma by Feature-Wise Attentional Graph Neural Network

PURPOSE

This study aimed to propose a regional lymph node (LN) metastasis prediction model for patients with esophageal squamous cell carcinoma (ESCC) that can learn and adaptively integrate preoperative computed tomography (CT) image features and nonimaging clinical parameters.

METHODS AND MATERIALS

Contrast-enhanced CT scans taken 2 weeks before surgery and 20 clinical factors, including general, pathologic, hematological, and diagnostic information, were collected from 357 patients with ESCC between October 2013 and November 2018. There were 999 regional LNs (857 negative, 142 positive) with pathologically confirmed status after surgery. All LNs were randomly divided into a training set (n = 738) and a validation set (n = 261) for testing. The feature-wise attentional graph neural network (FAGNN) was composed of (1) deep image feature extraction by the encoder of 3-dimensional UNet and high-level nonimaging factor representation by the clinical parameter encoder; (2) a feature-wise attention module for feature embedding with learnable adaptive weights; and (3) a graph attention layer to integrate the embedded features for final LN level metastasis prediction.

RESULTS

Among the 4 models we constructed, FAGNN using both CT and clinical parameters as input is the model with the best performance, and the area under the curve (AUC) reaches 0.872, which is better than manual CT diagnosis method, multivariable model using CT only (AUC = 0.797), multivariable model with combined CT and clinical parameters (AUC = 0.846), and our FAGNN using CT only (AUC = 0.853).

CONCLUSIONS

Our adaptive integration model improved the metastatic LN prediction performance based on CT and clinical parameters. Our model has the potential to foster effective fusion of multisourced parameters and to support early prognosis and personalized surgery or radiation therapy planning in patients with ESCC.

Feature selection methods and predictive models in CT lung cancer radiomics

Radiomics is a technique that extracts quantitative features from medical images using data-characterization algorithms. Radiomic features can be used to identify tissue characteristics and radiologic phenotyping that is not observable by clinicians. A typical workflow for a radiomics study includes cohort selection, radiomic feature extraction, feature and predictive model selection, and model training and validation. While there has been increasing attention given to radiomic feature extraction, standardization, and reproducibility, currently, there is a lack of rigorous evaluation of feature selection methods and predictive models. Herein, we review the published radiomics investigations in CT lung cancer and provide an overview of the commonly used radiomic feature selection methods and predictive models. We also compare limitations of various methods in clinical applications and present sources of uncertainty associated with those methods. This review is expected to help raise awareness of the impact of radiomic feature and model selection methods on the integrity of radiomics studies.

Value of Presurgical 18F-FDG PET/CT Radiomics for Predicting Mediastinal Lymph Node Metastasis in Patients with Lung Adenocarcinoma

Objective: The aim of this study was to develop an F-fluorodeoxyglucose (18F-FDG) positron emission tomography/computed tomography (PET/CT) radiomic model for predicting mediastinal lymph node metastasis (LNM) in presurgical patients with lung adenocarcinoma. Methods: The study enrolled 320 patients with lung adenocarcinoma (288 internal and 32 external cases) and extracted 190 radiomic features using the LIFEx package. Optimal radiomic features to build a radiomic model were selected using the least absolute shrinkage and selection operator algorithm. Logistic regression was used to build the clinical and complex (combined radiomic and clinical variables) models. Results: Ten radiomic features were selected. In the training group, the area under the receiver operating characteristic curve of the complex model was significantly higher than that of the radiomic and clinical models [0.924 (95% CI: 0.887-0.961) vs. 0.863 (95% CI: 0.814-0.912; p = 0.001) and 0.838 (95% CI: 0.783-0.894; p = 0.000), respectively]. The sensitivity, specificity, accuracy, and positive and negative predictive values of the radiomic model were 0.857, 0.790, 0.811, and 0.651 and 0.924, respectively, which were better than that of visual evaluation (0.539, 0.724, 0.667, and 0.472 and 0.775, respectively) and PET semiquantitative analyses (0.619, 0.732, 0.697, and 0.513 and 0.808, respectively). Conclusions: 18F-FDG PET/CT radiomics showed good predictive performance for LNM and improved the N-stage accuracy of lung adenocarcinoma.

Prediction of Radiation-Induced Hypothyroidism Using Radiomic Data Analysis Does Not Show Superiority over Standard Normal Tissue Complication Models

Simple Summary

Radiation-induced hypothyroidism (RIHT) commonly develops in cancer survivors that receive radiation therapy for cancers in the head and neck region. The state-of-art normal tissue complication probability (NTCP) models perform satisfactorily; however, they do not use the whole spectrum of information that can be obtained from imaging techniques. The radiomic approach offers the ability to efficiently mine features, which are imperceptible to the human eye, but may provide crucial data about the patient’s condition. We gathered CT images and clinical data from 98 patients undergoing radiotherapy for head and neck cancers, 27 of whom later developed RIHT. For them, we created machine-learning models to predict RIHT using automatically extracted radiomic features and appropriate clinical and dosimetric parameters. We also validated the well-established external state-of-art NTCP models on our datasets and observed that our radiomic-based models performed very similarly to them. This shows that automated tools may perform as well as the current standard but can be theoretically applied faster and be implemented into existing imaging software used when planning radiotherapy.

Abstract

State-of-art normal tissue complication probability (NTCP) models do not take into account more complex individual anatomical variations, which can be objectively quantitated and compared in radiomic analysis. The goal of this project was development of radiomic NTCP model for radiation-induced hypothyroidism (RIHT) using imaging biomarkers (radiomics). We gathered CT images and clinical data from 98 patients, who underwent intensity-modulated radiation therapy (IMRT) for head and neck cancers with a planned total dose of 70.0 Gy (33–35 fractions). During the 28-month (median) follow-up 27 patients (28%) developed RIHT. For each patient, we extracted 1316 radiomic features from original and transformed images using manually contoured thyroid masks. Creating models based on clinical, radiomic features or a combination thereof, we considered 3 variants of data preprocessing. Based on their performance metrics (sensitivity, specificity), we picked best models for each variant ((0.8, 0.96), (0.9, 0.93), (0.9, 0.89) variant-wise) and compared them with external NTCP models ((0.82, 0.88), (0.82, 0.88), (0.76, 0.91)). We showed that radiomic-based models did not outperform state-of-art NTCP models (p > 0.05). The potential benefit of radiomic-based approach is that it is dose-independent, and models can be used prior to treatment planning allowing faster selection of susceptible population.

Computed tomography-based radiomic model at node level for the prediction of normal-sized lymph node metastasis in cervical cancer

•

The metastatic status of lymph nodes in cervical cancer patients can be predicted.

•

Computed tomography-based radiomic model can identify the status of the normal-sized lymph node singly.

•

The model may help doctors to make staging and clinical decision, and realize individualized treatment.

Purpose

Radiomic models have been demonstrated to have acceptable discrimination capability for detecting lymph node metastasis (LNM). We aimed to develop a computed tomography–based radiomic model and validate its usefulness in the prediction of normal-sized LNM at node level in cervical cancer.

Methods

A total of 273 LNs of 219 patients from 10 centers were evaluated in this study. We randomly divided the LNs from the 2 centers with the largest number of LNs into the training and internal validation cohorts, and the rest as the external validation cohort. Radiomic features were extracted from the arterial and venous phase images. We trained an artificial neural network (ANN) to develop two single-phase models. A radiomic model reflecting the features of two-phase images was also built for directly predicting LNM in cervical cancer. Moreover, four state-of-the-art methods were used for comparison. The performance of all models was assessed using the area under the receiver operating characteristic curve (AUC).

Results

Among the models we built, the models combining the features of two phases surpassed the single-phase models, and the models generated by ANN had better performance than the others. We found that the radiomic model achieved the highest AUCs of 0.912 and 0.859 in the training and internal validation cohorts, respectively. In the external validation cohort, the AUC of the radiomic model was 0.800.

Conclusion

We constructed a radiomic model that exhibited great ability in the prediction of LNM. The application of the model could optimize clinical staging and decision-making.

Prognostic Value of Deep Learning-Mediated Treatment Monitoring in Lung Cancer Patients Receiving Immunotherapy

BACKGROUND

Checkpoint inhibitors provided sustained clinical benefit to metastatic lung cancer patients. Nonetheless, prognostic markers in metastatic settings are still under research. Imaging offers distinctive advantages, providing whole-body information non-invasively, while routinely available in most clinics. We hypothesized that more prognostic information can be extracted by employing artificial intelligence (AI) for treatment monitoring, superior to 2D tumor growth criteria.

METHODS

A cohort of 152 stage-IV non-small-cell lung cancer patients (NSCLC) (73 discovery, 79 test, 903CTs), who received nivolumab were retrospectively collected. We trained a neural network to identify morphological changes on chest CT acquired during patients' follow-ups. A classifier was employed to link imaging features learned by the network with overall survival.

RESULTS

Our results showed significant performance in the independent test set to predict 1-year overall survival from the date of image acquisition, with an average area under the curve (AUC) of 0.69 (p < 0.01), up to AUC 0.75 (p < 0.01) in the first 3 to 5 months of treatment, and 0.67 AUC (p = 0.01) for durable clinical benefit (6 months progression-free survival). We found the AI-derived survival score to be independent of clinical, radiological, PDL1, and histopathological factors. Visual analysis of AI-generated prognostic heatmaps revealed relative prognostic importance of morphological nodal changes in the mediastinum, supraclavicular, and hilar regions, lung and bone metastases, as well as pleural effusions, atelectasis, and consolidations.

CONCLUSIONS

Our results demonstrate that deep learning can quantify tumor- and non-tumor-related morphological changes important for prognostication on serial imaging. Further investigation should focus on the implementation of this technique beyond thoracic imaging.

Preoperatively Estimating the Malignant Potential of Mediastinal Lymph Nodes: A Pilot Study Toward Establishing a Robust Radiomics Model Based on Contrast-Enhanced CT Imaging

PURPOSE

To establish and validate a radiomics model to estimate the malignancy of mediastinal lymph nodes (LNs) based on contrast-enhanced CT imaging.

METHOD

In total, 201 pathologically confirmed mediastinal LNs from 129 patients were enrolled and assigned to training and test sets. Radiomics features were extracted from the region of interest (ROI) delineated on venous-phase CT imaging of LN. Feature selection was performed with least absolute shrinkage and selection operator (LASSO) binary logistic regression. Multivariate logistic regression was performed with the backward stepwise elimination. A model was fitted to associate mediastinal LN malignancy with selected features. The performance of the model was assessed and compared to that of five other machine learning algorithms (support vector machine, naive Bayes, random forest, decision tree, K-nearest neighbor) using receiver operating characteristic (ROC) curves. Calibration curves and Hosmer-Lemeshow tests were used to assess the calibration degree. Decision curve analysis (DCA) was used to assess the clinical usefulness of the logistic regression model in both the training and test sets. Stratified analysis was performed for different scanners and slice thicknesses.

RESULT

Among the six machine learning methods, the logistic regression model with the eight strongest features showed a significant association with mediastinal LN status and the satisfactory diagnostic performance for distinguishing malignant LNs from benign LNs. The accuracy, sensitivity, specificity and area under the ROC curve (AUC) were 0.850/0.803, 0.821/0.806, 0.893/0.800, and 0.922/0.850 in the training/test sets, respectively. The Hosmer-Lemeshow test showed that the P value was > 0.05, indicating good calibration, and the calibration curves showed good agreement between the classifications and actual observations. DCA showed that the model would obtain more benefit when the threshold probability was between 30% and 90% in the test set. Stratified analysis showed that the performance was not affected by different scanners or slice thicknesses. There was no significant difference (DeLong test, P > 0.05) between any two subgroups, which showed the generalization of the radiomics score across different factors.

CONCLUSION

The model we built could help assist the preoperative estimation of mediastinal LN malignancy based on contrast-enhanced CT imaging, with stability for different scanners and slice thicknesses.