Investigation of the added value of CT-based radiomics in predicting the development of brain metastases in patients with radically treated stage III NSCLC

Abstract

Introduction: Despite radical intent therapy for patients with stage III non-small-cell lung cancer (NSCLC), cumulative incidence of brain metastases (BM) reaches 30%. Current risk stratification methods fail to accurately identify these patients. As radiomics features have been shown to have predictive value, this study aims to develop a model combining clinical risk factors with radiomics features for BM development in patients with radically treated stage III NSCLC. Methods: Retrospective analysis of two prospective multicentre studies. Inclusion criteria: adequately staged [18F-fluorodeoxyglucose positron emission tomography-computed tomography (18-FDG-PET-CT), contrast-enhanced chest CT, contrast-enhanced brain magnetic resonance imaging/CT] and radically treated stage III NSCLC, exclusion criteria: second primary within 2 years of NSCLC diagnosis and prior prophylactic cranial irradiation. Primary endpoint was BM development any time during follow-up (FU). CT-based radiomics features ( N = 530) were extracted from the primary lung tumour on 18-FDG-PET-CT images, and a list of clinical features ( N = 8) was collected. Univariate feature selection based on the area under the curve (AUC) of the receiver operating characteristic was performed to identify relevant features. Generalized linear models were trained using the selected features, and multivariate predictive performance was assessed through the AUC. Results: In total, 219 patients were eligible for analysis. Median FU was 59.4 months for the training cohort and 67.3 months for the validation cohort; 21 (15%) and 17 (22%) patients developed BM in the training and validation cohort, respectively. Two relevant clinical features (age and adenocarcinoma histology) and four relevant radiomics features were identified as predictive. The clinical model yielded the highest AUC value of 0.71 (95% CI: 0.58–0.84), better than radiomics or a combination of clinical parameters and radiomics (both an AUC of 0.62, 95% CIs of 0.47–076 and 0.48–0.76, respectively). Conclusion: CT-based radiomics features of primary NSCLC in the current setup could not improve on a model based on clinical predictors (age and adenocarcinoma histology) of BM development in radically treated stage III NSCLC patients.

Risk factors for brain metastases in locally advanced non-small cell lung cancer patients treated with radical radiotherapy

“…It has been widely reported that patients with non-SCC or adenocarcinoma had a higher risk for developing BM than patients with SCC ( 9 , 15 - 18 ). We confirmed that non-SCC was an independent predictive factor for BM in our study (P<0.001; HR: 6.08; 95% CI: 2.26–16.37).…”

A cuproptosis score model and prognostic score model can evaluate clinical characteristics and immune microenvironment in NSCLC

“…However, the relative survival rate in advanced NSCLC is still approximately 15% [ 7 ]. Despite computed tomography (CT) has reduced the mortality rate of early NSCLC by 20% [ 8 ], and the application of PET-CT has greatly improved the precision of NSCLC grading, merely depending on imaging methods alone are still unable to more accurately assess the development or prognosis of NSCLC patients and choice of treatment strategy [ 9 , 10 ]. Hence, establishing a viable biomarkers and prognostic models, thereby revealing the prognostic genetic features of NSCLC in clinical practice is of considerable significance.…”

Enhancing brain metastasis prediction in non-small cell lung cancer: a deep learning-based segmentation and CT radiomics-based ensemble learning model

“…The BM status of advanced NSCLC patients may influence the treatment efficacy of chemotherapy, immunotherapy, and radiotherapy [ 4 – 7 ]. Previous studies have proven that prophylactic cranial irradiation (PCI) is an effective way to prevent the morbidity associated with BM in NSCLC patients [ 8 ]. Although PCI can reduce the occurrence rate of BM by approximately 50%, it fails to improve overall survival (OS) of NSCLC patients.…”

“…The results indicated that the prediction performance in terms of predicting BM-free survival in curatively resected LA-NSCLC patients can be improved by integrating CT radiomics and clinical features. However, another study provided an in-depth comparative analysis of CT-based radiomics and clinical factors and showed that CT-based radiomics features of primary NSCLC cannot improve the predictive efficiency of a clinical risk factor (age and adenocarcinoma histology)-based model for BM development in radically treated stage III NSCLC patients [ 8 ]. The contradictory results yielded by these studies may be caused by the different populations of enrolled patients.…”