Pre-Immunotherapy Contrast-Enhanced CT Texture-Based Classification: A Useful Approach to Non-Small Cell Lung Cancer Immunotherapy Efficacy Prediction

Multiomics-Based Deep Learning Prediction of Prognosis and Therapeutic Response in Patients With Extensive-Stage Small Cell Lung Cancer Receiving Chemoimmunotherapy: A Retrospective Cohort Study

Objective

This study aimed to develop a clinical early warning prediction model to evaluate the prognosis and response to chemoimmunotherapy in patients with extensive-stage small cell lung cancer (ES-SCLC), thereby guiding clinical decision-making.

Methods

A retrospective analysis was conducted on the clinical data and radiomics parameters of 309 patients with ES-SCLC hospitalized at Baotou Cancer Hospital from February 2020 to September 2024. Patients were divided into reactive and non-reactive groups based on their response to chemoimmunotherapy.Machine learning algorithms (including random forests, decision trees, artificial neural networks, and generalized linear regression) were used to predict the combined treatment response. The model's predictive ability was evaluated using the receiver operating characteristic (ROC) curve and clinical decision curve analysis(DCA). The prognostic evaluation of patients receiving combination therapy was based on the COX regression model, with predictive performance assessed through nomogram visualization and calibration curves.

Results

Out of 309 patients with ES-SCLC, 248 (80.26%) responded to combination therapy. Logistic regression and Least absolute shrinkage and selection operator (LASSO) regression analyses identified Energy, sum of squares(SOS), mean sum(MES), sum variance(SUV), sum entropy(SUE), difference variance(DIV), and pathomics score as independent risk factors for treatment response. The area under the ROC curve for predicting treatment response using machine learning were 0.764 (95% confidence interval [CI]: 0.707~0.821) and 0.901 (95% CI: 0.846~0.956) in the training and validation sets. The C-index of the radiomics and pathomics prognostic nomogram model based on the COX prognostic model was 0.766 and 0.812 in those sets, respectively.

Conclusion

We developed prediction model based on multi-omics demonstrated satisfactory performance in predicting chemoimmunotherapy response in patients with ES-SCLC. The random forest prediction model, in particular, provides accurate response and prognostic risk assessments, thereby assisting clinical decision-making.

Automated Lung Cancer Diagnosis Applying Butterworth Filtering, Bi-Level Feature Extraction, and Sparce Convolutional Neural Network to Luna 16 CT Images

Accurate prognosis and diagnosis are crucial for selecting and planning lung cancer treatments. As a result of the rapid development of medical imaging technology, the use of computed tomography (CT) scans in pathology is becoming standard practice. An intricate interplay of requirements and obstacles characterizes computer-assisted diagnosis, which relies on the precise and effective analysis of pathology images. In recent years, pathology image analysis tasks such as tumor region identification, prognosis prediction, tumor microenvironment characterization, and metastasis detection have witnessed the considerable potential of artificial intelligence, especially deep learning techniques. In this context, an artificial intelligence (AI)-based methodology for lung cancer diagnosis is proposed in this research work. As a first processing step, filtering using the Butterworth smooth filter algorithm was applied to the input images from the LUNA 16 lung cancer dataset to remove noise without significantly degrading the image quality. Next, we performed the bi-level feature selection step using the Chaotic Crow Search Algorithm and Random Forest (CCSA-RF) approach to select features such as diameter, margin, spiculation, lobulation, subtlety, and malignancy. Next, the Feature Extraction step was performed using the Multi-space Image Reconstruction (MIR) method with Grey Level Co-occurrence Matrix (GLCM). Next, the Lung Tumor Severity Classification (LTSC) was implemented by using the Sparse Convolutional Neural Network (SCNN) approach with a Probabilistic Neural Network (PNN). The developed method can detect benign, normal, and malignant lung cancer images using the PNN algorithm, which reduces complexity and efficiently provides classification results. Performance parameters, namely accuracy, precision, F-score, sensitivity, and specificity, were determined to evaluate the effectiveness of the implemented hybrid method and compare it with other solutions already present in the literature.

Defining clinically useful biomarkers of immune checkpoint inhibitors in solid tumours

Although more than a decade has passed since the approval of immune checkpoint inhibitors (ICIs) for the treatment of melanoma and non-small-cell lung, breast and gastrointestinal cancers, many patients still show limited response. US Food and Drug Administration (FDA)-approved biomarkers include programmed cell death 1 ligand 1 (PDL1) expression, microsatellite status (that is, microsatellite instability-high (MSI-H)) and tumour mutational burden (TMB), but these have limited utility and/or lack standardized testing approaches for pan-cancer applications. Tissue-based analytes (such as tumour gene signatures, tumour antigen presentation or tumour microenvironment profiles) show a correlation with immune response, but equally, these demonstrate limited efficacy, as they represent a single time point and a single spatial assessment. Patient heterogeneity as well as inter- and intra-tumoural differences across different tissue sites and time points represent substantial challenges for static biomarkers. However, dynamic biomarkers such as longitudinal biopsies or novel, less-invasive markers such as blood-based biomarkers, radiomics and the gut microbiome show increasing potential for the dynamic identification of ICI response, and patient-tailored predictors identified through neoadjuvant trials or novel ex vivo tumour models can help to personalize treatment. In this Perspective, we critically assess the multiple new static, dynamic and patient-specific biomarkers, highlight the newest consortia and trial efforts, and provide recommendations for future clinical trials to make meaningful steps forwards in the field.

Genomic alterations associated with pseudoprogression and hyperprogressive disease during anti-PD1 treatment for advanced non-small-cell lung cancer

Introduction

This study aimed to elucidate the relationship between dynamic genomic mutation alteration and pseudoprogression (PsPD)/hyperprogressive disease (HPD) in immunotherapy-treated advanced non-small-cell lung cancer (NSCLC), to provide clinical evidence for identifying and distinguishing between PsPD and HPD.

Method

Patients with advanced NSCLC who were treated with anti-PD1 were enrolled. Whole blood was collected at baseline and post image progression. Serum was separated and sequenced using 425-panel next-generation sequencing analysis (NGS).

Results

NGS revealed that not only single gene mutations were associated with PsPD/HPD before treatment, dynamic monitoring of the whole-blood genome mutation spectrum also varied greatly. Mutational burden, allele frequency%, and relative circulating tumor DNA abundance indicated that the fold change after image progression was much higher in the HPD group.

Discussion

The gene mutation profiles of PsPD and HPD not only differed before treatment, but higher genome mutation spectrum post image progression indicated true disease progression in patients with HPD. This suggests that dynamic whole-genome mutation profile monitoring as NGS can distinguish PsPD from HPD more effectively than single gene detection, providing a novel method for guiding clinical immune treatment.

Immunotherapy Efficacy Prediction for Non-Small Cell Lung Cancer Using Multi-View Adaptive Weighted Graph Convolutional Networks

Immunotherapy is an effective way to treat non-small cell lung cancer (NSCLC). The efficacy of immunotherapy differs from person to person and may cause side effects, making it important to predict the efficacy of immunotherapy before surgery. Radiomics based on machine learning has been successfully used to predict the efficacy of NSCLC immunotherapy. However, most studies only considered the radiomic features of the individual patient, ignoring the inter-patient correlations. Besides, they usually concatenated different features as the input of a single-view model, failing to consider the complex correlation among features of multiple types. To this end, we propose a multi-view adaptive weighted graph convolutional network (MVAW-GCN) for the prediction of NSCLC immunotherapy efficacy. Specifically, we group the radiomic features into several views according to the type of the fitered images they extracted from. We construct a graph in each view based on the radiomic features and phenotypic information. An attention mechanism is introduced to automatically assign weights to each view. Considering the view-shared and view-specific knowledge of radiomic features, we propose separable graph convolution that decomposes the output of the last convolution layer into two components, i.e., the view-shared and view-specific outputs. We maximize the consistency and enhance the diversity among different views in the learning procedure. The proposed MVAW-GCN is evaluated on 107 NSCLC patients, including 52 patients with valid efficacy and 55 patients with invalid efficacy. Our method achieved an accuracy of 77.27% and an area under the curve (AUC) of 0.7780, indicating its effectiveness in NSCLC immunotherapy efficacy prediction.

Noninvasive assessment of significant liver fibrosis in rabbits by spectral CT parameters and texture analysis

PURPOSE

Noninvasive assessment of significant liver fibrosis in rabbits by spectral CT parameters and texture analysis.

MATERIALS AND METHODS

Thirty-three rabbits were randomly divided into 27 carbon tetrachloride-induced liver fibrosis group and 6 control group. Spectral CT contrast-enhanced scan was performed in batches, and the liver fibrosis was staged according to the histopathological results. The portal venous phase spectral CT parameters [70 keV CT value, normalized iodine concentration (NIC), spectral HU curve slope (λHU)] were measured, and MaZda texture analysis was performed on 70 keV monochrome images. Three dimensionality reduction methods and four statistical methods in B11 module were used to perform discriminant analysis and calculate misclassified rate (MCR), and ten texture features under the lowest combination of MCR were statistically analyzed. Receiver operating characteristic curve (ROC) was used to calculate the diagnostic performance of spectral parameters and texture features for significant liver fibrosis. Finally, the binary logistic regression was used to further screen independent predictors and establish model.

RESULTS

A total of 23 experimental rabbits and 6 control rabbits were included, of which 16 had significant liver fibrosis. Three spectral CT parameters with significant liver fibrosis were significantly lower than those of non-significant liver fibrosis (p < 0.05), and the AUC ranged from 0.846 to 0.913. The combination analysis of mutual information (MI) and nonlinear discriminant analysis (NDA) had the lowest MCR, which with 0%. In the filtered texture features, four were statistically significant and AUC > 0.5, ranges from 0.764 to 0.875. The logistic regression model showed that Perc.90% and NIC could be used as independent predictors, the overall prediction accuracy of the model was 89.7% and the AUC was 0.976.

CONCLUSION

Spectral CT parameters and texture features have high diagnostic value for predicting significant liver fibrosis in rabbits, and the combination of the two can improve its diagnostic efficiency.

Radiomic biomarkers from chest computed tomography are assistive in immunotherapy response prediction for non-small cell lung cancer

Background

Immunotherapies, such as programmed death 1/programmed death ligand 1 (PD-1/PD-L1) antibodies have been shown to improve overall and progression-free survival (PFS) in patients with locally advanced or metastatic non-small cell lung cancer (NSCLC). However, not all patients derive a meaningful clinical benefit. Additionally, patients receiving anti-PD-1/PD-L1 therapy can experience immune-related adverse events (irAEs). Clinically significant irAEs may require temporary pause or discontinuation of treatment. Having a tool to identify patients who may not benefit and/or are at risk for developing severe irAEs from immunotherapy will aid in an informed decision-making process for the patients and their physicians.

Methods

Computed tomography (CT) scans and clinical data were retrospectively collected for this study to develop three prediction models using (I) radiomic features, (II) clinical features, and (III) radiomic and clinical features combined. Each subject had 6 clinical features and 849 radiomic features extracted. Selected features were run through an artificial neural network (NN) trained on 70% of the cohort, maintaining the case and control ratio. The NN was assessed by calculating the area-under-the-receiver-operating-characteristic curve (AUC-ROC), area-under-the-precision-recall curve (AUC-PR), sensitivity, and specificity.

Results

A cohort of 132 subjects, of which 43 (33%) had a PFS ≤90 days and 89 (67%) of which had a PFS >90 days was used to develop the prediction models. The radiomic model was able to predict progression-free survival with a training AUC-ROC of 87% and testing AUC-ROC, sensitivity, and specificity of 83%, 75%, and 81%, respectively. In this cohort, the clinical and radiomic combined features did add a slight increase in the specificity (85%) but with a decrease in sensitivity (75%) and AUC-ROC (81%).

Conclusions

Whole lung segmentation and feature extraction can identify those that would see a benefit from anti-PD-1/PD-L1 therapy.

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.

CT-based radiomics in predicting pathological response in non-small cell lung cancer patients receiving neoadjuvant immunotherapy

Objectives

In radiomics, high-throughput algorithms extract objective quantitative features from medical images. In this study, we evaluated CT-based radiomics features, clinical features, in-depth learning features, and a combination of features for predicting a good pathological response (GPR) in non-small cell lung cancer (NSCLC) patients receiving immunotherapy-based neoadjuvant therapy (NAT).

Materials and methods

We reviewed 62 patients with NSCLC who received surgery after immunotherapy-based NAT and collected clinicopathological data and CT images before and after immunotherapy-based NAT. A series of image preprocessing was carried out on CT scanning images: tumor segmentation, conventional radiomics feature extraction, deep learning feature extraction, and normalization. Spearman correlation coefficient, principal component analysis (PCA), and least absolute shrinkage and selection operator (LASSO) were used to screen features. The pretreatment traditional radiomics combined with clinical characteristics (before_rad_cil) model and pretreatment deep learning characteristics (before_dl) model were constructed according to the data collected before treatment. The data collected after NAT created the after_rad_cil model and after_dl model. The entire model was jointly constructed by all clinical features, conventional radiomics features, and deep learning features before and after neoadjuvant treatment. Finally, according to the data obtained before and after treatment, the before_nomogram and after_nomogram were constructed.

Results

In the before_rad_cil model, four traditional radiomics features ("original_shape_flatness," "wavelet hhl_firer_skewness," "wavelet hlh_firer_skewness," and "wavelet lll_glcm_correlation") and two clinical features ("gender" and "N stage") were screened out to predict a GPR. The average prediction accuracy (ACC) after modeling with k-nearest neighbor (KNN) was 0.707. In the after_rad_cil model, nine features predictive of GPR were obtained after feature screening, among which seven were traditional radiomics features: "exponential_firer_skewness," "exponential_glrlm_runentropy," "log- sigma-5-0-mm-3d_firer_kurtosis," "logarithm_skewness," "original_shape_elongation," "original_shape_brilliance," and "wavelet llh_glcm_clustershade"; two were clinical features: "after_CRP" and "after lymphocyte percentage." The ACC after modeling with support vector machine (SVM) was 0.682. The before_dl model and after_dl model were modeled by SVM, and the ACC was 0.629 and 0.603, respectively. After feature screening, the entire model was constructed by multilayer perceptron (MLP), and the ACC of the GPR was the highest, 0.805. The calibration curve showed that the predictions of the GPR by the before_nomogram and after_nomogram were in consensus with the actual GPR.

Conclusion

CT-based radiomics has a good predictive ability for a GPR in NSCLC patients receiving immunotherapy-based NAT. Among the radiomics features combined with the clinicopathological information model, deep learning feature model, and the entire model, the entire model had the highest prediction accuracy.

Radiomics based on enhanced CT for differentiating between pulmonary tuberculosis and pulmonary adenocarcinoma presenting as solid nodules or masses

PURPOSE

To investigate the incremental value of enhanced CT-based radiomics in discriminating between pulmonary tuberculosis (PTB) and pulmonary adenocarcinoma (PAC) presenting as solid nodules or masses and to develop an optimal radiomics model.

METHODS

A total of 128 lesions (from 123 patients) from three hospitals were retrospectively analyzed and were randomly divided into training and test datasets at a ratio of 7:3. Independent predictors in subjective image features were used to develop the subjective image model (SIM). The plain CT-based and enhanced CT-based radiomics features were screened by the correlation coefficient method, univariate analysis, and the least absolute shrinkage and selection operator, then used to build the plain CT radiomics model (PRM) and enhanced CT radiomics model (ERM), respectively. Finally, the combined model (CM) combining PRM and ERM was established. In addition, the performance of three radiologists and one respiratory physician was evaluated. The areas under the receiver operating characteristic curve (AUCs) were used to assess the performance of each model.

RESULTS

The differential diagnostic capability of the ERM (training: AUC = 0.933; test: AUC = 0.881) was better than that of the PRM (training: AUC = 0.861; test: AUC = 0.756) and the SIM (training: AUC = 0.760; test: AUC = 0.611). The CM was optimal (training: AUC = 0.948; test: AUC = 0.917) and outperformed the respiratory physician and most radiologists.

CONCLUSIONS

The ERM was more helpful than the PRM for identifying PTB and PAC that present as solid nodules or masses, and the CM was the best.

The Role of Radiomics in the Era of Immune Checkpoint Inhibitors: A New Protagonist in the Jungle of Response Criteria

Simple Summary

The introduction of immune checkpoint inhibitors has represented a milestone in cancer treatment. Despite PD-L1 expression being the standard biomarker used before the start of therapy, there is still a strict need to identify complementary non-invasive biomarkers in order to better select patients. In this context, radiomics is an emerging approach for examining medical images and clinical data by capturing multiple features hidden from human eye and is potentially able to predict response assessment and survival in the course of immunotherapy. We reviewed the available studies investigating the role of radiomics in cancer patients, focusing on non-small cell lung cancer treated with immune checkpoint inhibitors. Although preliminary research shows encouraging results, different issues need to be solved before radiomics can enter into clinical practice.

Abstract

Immune checkpoint inhibitors (ICI) have demonstrated encouraging results in terms of durable clinical benefit and survival in several malignancies. Nevertheless, the search to identify an “ideal” biomarker for predicting response to ICI is still far from over. Radiomics is a new translational field of study aiming to extract, by dedicated software, several features from a given medical image, ranging from intensity distribution and spatial heterogeneity to higher-order statistical parameters. Based on these premises, our review aims to summarize the current status of radiomics as a potential predictor of clinical response following immunotherapy treatment. A comprehensive search of PubMed results was conducted. All studies published in English up to and including December 2021 were selected, comprising those that explored computed tomography (CT), magnetic resonance imaging (MRI), and positron emission tomography (PET) for radiomic analyses in the setting of ICI. Several studies have demonstrated the potential applicability of radiomic features in the monitoring of the therapeutic response beyond the traditional morphologic and metabolic criteria, as well as in the prediction of survival or non-invasive assessment of the tumor microenvironment. Nevertheless, important limitations emerge from our review in terms of standardization in feature selection, data sharing, and methods, as well as in external validation. Additionally, there is still need for prospective clinical trials to confirm the potential significant role of radiomics during immunotherapy.