Predicting brain metastasis in early stage non-small cell lung cancer patients by gene expression profiling

Machine-Learning-Aided Prediction of Brain Metastases Development in Non-Small-Cell Lung Cancers

PURPOSE

Non-small-cell lung cancer (NSCLC) shows a high incidence of brain metastases (BM). Early detection is crucial to improve clinical prospects. We trained and validated classifier models to identify patients with a high risk of developing BM, as they could potentially benefit from surveillance brain MRI.

METHODS

Consecutive patients with an initial diagnosis of NSCLC from January 2011 to April 2019 and an in-house chest-CT scan (staging) were retrospectively recruited at a German lung cancer center. Brain imaging was performed at initial diagnosis and in case of neurological symptoms (follow-up). Subjects lost to follow-up or still alive without BM at the data cut-off point (12/2020) were excluded. Covariates included clinical and/or 3D-radiomics-features of the primary tumor from staging chest-CT. Four machine learning models for prediction (80/20 training) were compared. Gini Importance and SHAP were used as measures of importance; sensitivity, specificity, area under the precision-recall curve, and Matthew's Correlation Coefficient as evaluation metrics.

RESULTS

Three hundred and ninety-five patients compromised the clinical cohort. Predictive models based on clinical features offered the best performance (tuned to maximize recall: sensitivity∼70%, specificity∼60%). Radiomics features failed to provide sufficient information, likely due to the heterogeneity of imaging data. Adenocarcinoma histology, lymph node invasion, and histological tumor grade were positively correlated with the prediction of BM, age, and squamous cell carcinoma histology were negatively correlated. A subgroup discovery analysis identified 2 candidate patient subpopulations appearing to present a higher risk of BM (female patients + adenocarcinoma histology, adenocarcinoma patients + no other distant metastases).

CONCLUSION

Analysis of the importance of input features suggests that the models are learning the relevant relationships between clinical features/development of BM. A higher number of samples is to be prioritized to improve performance. Employed prospectively at initial diagnosis, such models can help select high-risk subgroups for surveillance brain MRI.

Noggin contributes to brain metastatic colonization of lung cancer cells

BACKGROUND

Brain metastasis is a common complication among patients with lung cancer, yet the underlying mechanisms remain unclear. In this study, we aimed to investigate the pathogenesis of brain metastasis in lung cancer.

METHODS

We established highly colonizing metastatic lung cancer cells, A549-M2, through multiple implantations of A549 human lung cancer cells in the carotid artery of athymic nude mice.

RESULTS

Compared to parental cells (M0), M2 cells demonstrated slower growth in culture plates and soft agar, as well as lower motility and higher adhesion, key characteristics of mesenchymal-epithelial transition (MET). Further analysis revealed that M2 cells exhibited decreased expression of epithelial-mesenchymal transition markers, including ZEB1 and Vimentin. M2 cells also demonstrated reduced invasiveness in co-culture systems. RNA sequencing and gene set enrichment analysis confirmed that M2 cells underwent MET. Intriguingly, depletion of Noggin, a BMP antagonist, was observed in M2 cells, and replenishment of Noggin restored suppressed migration and invasion of M2 cells. In addition, Noggin knockdown in control M0 cells promoted cell attachment and suppressed cell migration, suggesting that Noggin reduction during brain colonization causes inhibition of migration and invasion of metastatic lung cancer cells.

CONCLUSIONS

Our results suggest that lung cancer cells undergo MET and lose their motility and invasiveness during brain metastatic colonization, which is dependent on Noggin.

Modern Stereotactic Radiotherapy for Brain Metastases from Lung Cancer: Current Trends and Future Perspectives Based on Integrated Translational Approaches

Brain metastases (BMs) represent the most frequent metastatic event in the course of lung cancer patients, occurring in approximately 50% of patients with non-small-cell lung cancer (NSCLC) and in up to 70% in patients with small-cell lung cancer (SCLC). Thus far, many advances have been made in the diagnostic and therapeutic procedures, allowing improvements in the prognosis of these patients. The modern approach relies on the integration of several factors, such as accurate histological and molecular profiling, comprehensive assessment of clinical parameters and precise definition of the extent of intracranial and extracranial disease involvement. The combination of these factors is pivotal to guide the multidisciplinary discussion and to offer the most appropriate treatment to these patients based on a personalized approach. Focal radiotherapy (RT), in all its modalities (radiosurgery (SRS), fractionated stereotactic radiotherapy (SRT), adjuvant stereotactic radiotherapy (aSRT)), is the cornerstone of BM management, either alone or in combination with surgery and systemic therapies. We review the modern therapeutic strategies available to treat lung cancer patients with brain involvement. This includes an accurate review of the different technical solutions which can be exploited to provide a "state-of-art" focal RT and also a detailed description of the systemic agents available as effective alternatives to SRS/SRT when a targetable molecular driver is present. In addition to the validated treatment options, we also discuss the future perspective for focal RT, based on emerging clinical reports (e.g., SRS for patients with many BMs from NSCLC or SRS for BMs from SCLC), together with a presentation of innovative and promising findings in translational research and the combination of novel targeted agents with SRS/SRT.

Advances in the Molecular Landscape of Lung Cancer Brain Metastasis

Lung cancer is one of the most frequent tumors that metastasize to the brain. Brain metastasis (BM) is common in advanced cases, being the major cause of patient morbidity and mortality. BMs are thought to arise via the seeding of circulating tumor cells into the brain microvasculature. In brain tissue, the interaction with immune cells promotes a microenvironment favorable to the growth of cancer cells. Despite multimodal treatments and advances in systemic therapies, lung cancer patients still have poor prognoses. Therefore, there is an urgent need to identify the molecular drivers of BM and clinically applicable biomarkers in order to improve disease outcomes and patient survival. The goal of this review is to summarize the current state of knowledge on the mechanisms of the metastatic spread of lung cancer to the brain and how the metastatic spread is influenced by the brain microenvironment, and to elucidate the molecular determinants of brain metastasis regarding the role of genomic and transcriptomic changes, including coding and non-coding RNAs. We also present an overview of the current therapeutics and novel treatment strategies for patients diagnosed with BM from NSCLC.

Artificial intelligence-based non-small cell lung cancer transcriptome RNA-sequence analysis technology selection guide

The incidence and mortality rates of lung cancer are high worldwide, where non-small cell lung cancer (NSCLC) accounts for more than 85% of lung cancer cases. Recent non-small cell lung cancer research has been focused on analyzing patient prognosis after surgery and identifying mechanisms in connection with clinical cohort and ribonucleic acid (RNA) sequencing data, including single-cell ribonucleic acid (scRNA) sequencing data. This paper investigates statistical techniques and artificial intelligence (AI) based non-small cell lung cancer transcriptome data analysis methods divided into target and analysis technology groups. The methodologies of transcriptome data were schematically categorized so researchers can easily match analysis methods according to their goals. The most widely known and frequently utilized transcriptome analysis goal is to find essential biomarkers and classify carcinomas and cluster NSCLC subtypes. Transcriptome analysis methods are divided into three major categories: Statistical analysis, machine learning, and deep learning. Specific models and ensemble techniques typically used in NSCLC analysis are summarized in this paper, with the intent to lay a foundation for advanced research by converging and linking the various analysis methods available.

Clinical features and molecular genetics associated with brain metastasis in suspected early-stage non-small cell lung cancer

Introduction

Regarding whether brain magnetic resonance imaging (MRI) should be routine in patients with suspected early-stage lung cancer, guideline recommendations are inconsistent. Therefore, we performed this study to evaluate the incidence of and risk factors for brain metastasis (BM) in patients with suspected early-stage non-small-cell lung cancer (NSCLC).

Methods

A review of the medical charts of consecutive NSCLC patients diagnosed between January 2006 and May 2020 was performed. We identified 1,382 NSCLC patients with clinical staging of T1/2aN0M0 (excluding BM), and investigated the incidence, clinical predictors, and prognosis of BM in the cohort. We also performed RNA-sequencing differential expression analysis using transcriptome of 8 patients, using DESeq2 package (version 1.32.0) with R (version 4.1.0).

Results

Among 1,382 patients, nine hundred forty-nine patients (68.7%) underwent brain MRI during staging, and 34 patients (3.6%) were shown to have BM. Firth's bias-reduced logistic regression showed that tumor size (OR 1.056; 95% CI 1.009-1.106, p=0.018) was the only predictor of BM, and pathologic type was not a predictor of BM in our cohort (p>0.05). The median overall survival for patients with brain metastasis was 5.5 years, which is better than previously reported in the literature. RNA-sequencing differential expression analysis revealed the top 10 significantly upregulated genes and top 10 significantly downregulated genes. Among the genes involved in BM, Unc-79 homolog, non-selective sodium leak channel (NALCN) channel complex subunit (UNC79) was the most highly expressed gene in the lung adenocarcinoma tissues from the BM group, and an in vitro assay using A549 cells revealed that the NALCN inhibitor suppressed lung cancer cell proliferation and migration.

Conclusions

Given the incidence and favorable outcome of BM in patients with suspected early-stage NSCLC, selective screening with brain MRI may be considered, especially in patients with high-risk features.

Brain metastases: It takes two factors for a primary cancer to metastasize to brain

Brain metastasis of a cancer is a malignant disease with high mortality, but the cause and the molecular mechanism remain largely unknown. Using the samples of primary tumors of 22 cancer types in the TCGA database, we have performed a computational study of their transcriptomic data to investigate the drivers of brain metastases at the basic physics and chemistry level. Our main discoveries are: (i) the physical characteristics, namely electric charge, molecular weight, and the hydrophobicity of the extracellular structures of the expressed transmembrane proteins largely affect a primary cancer cell’s ability to cross the blood-brain barrier; and (ii) brain metastasis may require specific functions provided by the activated enzymes in the metastasizing primary cancer cells for survival in the brain micro-environment. Both predictions are supported by published experimental studies. Based on these findings, we have built a classifier to predict if a given primary cancer may have brain metastasis, achieving the accuracy level at AUC = 0.92 on large test sets.

Emerging Biomarkers for Diagnosis, Prevention and Treatment of Brain Metastases-From Biology to Clinical Utility

Primary malignancies of the lung, skin (melanoma), and breast have higher propensity for metastatic spread to the brain. Advances in molecular tumour profiling have aided the development of targeted therapies, stereotactic radiotherapy, and immunotherapy, which have led to some improvement in patient outcomes; however, the overall prognosis remains poor. Continued research to identify new prognostic and predictive biomarkers is necessary to further impact patient outcomes, as this will enable better risk stratification at the point of primary cancer diagnosis, earlier detection of metastatic deposits (for example, through surveillance), and more effective systemic treatments. Brain metastases exhibit considerable inter- and intratumoural heterogeneity-apart from distinct histology, treatment history and other clinical factors, the metastatic brain tumour microenvironment is incredibly variable both in terms of subclonal diversity and cellular composition. This review discusses emerging biomarkers; specifically, the biological context and potential clinical utility of tumour tissue biomarkers, circulating tumour cells, extracellular vesicles, and circulating tumour DNA.

The Development of Machine Learning Methods in discriminating Secretory Proteins of Malaria Parasite

Malaria caused by Plasmodium falciparum is one of the major infectious diseases in the world. It is essential to exploit an effective method to predict secretory proteins of malaria parasites to develop effective cures and treatment. Biochemical assays can provide details for accurate identification of the secretory proteins, but these methods are expensive and time-consuming. In this paper, we summarized the machine learning-based identification algorithms and compared the construction strategies between different computational methods. Also, we discussed the use of machine learning to improve the ability of algorithms to identify proteins secreted by malaria parasites.

Cancer driver gene and non-coding RNA alterations as biomarkers of brain metastasis in lung cancer: A review of the literature

Brain metastasis (BM) is the most common event in patients with lung cancer. Despite multimodal treatments and advances in systemic therapies, development of BM remains one of the main factors associated with poor prognosis and mortality in patients with lung cancer. Therefore, better understanding of mechanisms involved in lung cancer brain metastasis (LCBM) is of great importance to suppress cancer cells and to improve the overall survival of patients. Several cancer-related genes such as EGFR and KRAS have been proposed as potential predictors of LCBM. In addition, there is ample evidence supporting crucial roles of non-coding RNAs (ncRNAs) in mediating LCBM. In this review, we provide comprehensive information on risk assessment, predictive, and prognostic panels for early detection of BM in patients with lung cancer. Moreover, we present an overview of LCBM molecular mechanisms, cancer driver genes, and ncRNAs which may predict the risk of BM in lung cancer patients. Recent clinical studies have focused on determining mechanisms involved in LCBM and their association with diagnosis, prognosis, and treatment outcomes. These studies have shown that alterations in EGFR, KRAS, BRAF, and ALK, as the most frequent coding gene alterations, and dysregulation of ncRNAs such as miR-423, miR-330-3p, miR-145, piR-651, and MALAT1 can be considered as potential biomarkers of LCBM.

Identification of a high-risk group for brain metastases in non-small cell lung cancer patients

PURPOSE

Identification of a high-risk group of brain metastases (BM) in patients with non-small cell lung cancer (NSCLC) could lead to early interventions and probably better prognosis. The objective of the study was to identify this group by generating a multivariable model with recognized and accessible risk factors.

METHODS

A retrospective cohort from patients seen at a single center during 2010-2020, was divided into a training (TD) and validation (VD) datasets, associations with BM were measured in the TD with logit, variables significantly associated were used to generate a multivariate model. Model´s performance was measured with the AUC/C-statistic, Akaike information criterion, and Brier score.

RESULTS

From 570 patients with NSCLC who met the strict eligibility criteria a TD and VD were randomly assembled, no significant differences were found amid both datasets. Variables associated with BM in the multivariate logit analyses were age [P 0.001, OR 0.96 (95% CI 0.93-0.98)]; mutational status positive [P 0.027, OR 1.96 (95% CI 1.07-3.56); and carcinoembryonic antigen levels [P 0.016, OR 1.001 (95% CI 1.000-1.003). BM were diagnosed in 24% of the whole cohort. Stratification into a high-risk group after simplification of the model, displayed a frequency of BM of 63% (P < 0.001).

CONCLUSION

A multivariate model comprising age, carcinoembryonic antigen levels, and mutation status allowed the identification of a truly high-risk group of BM in NSCLC patients.

Incidence and real-world burden of brain metastases from solid tumors and hematologic malignancies in Ontario: a population-based study

BACKGROUND

Although intracranial metastatic disease (IMD) is a frequent complication of cancer, most cancer registries do not capture these cases. Consequently, a data-gap exists, which thwarts system-level quality improvement efforts. The purpose of this investigation was to determine the real-world burden of IMD.

METHODS

Patients diagnosed with a non-CNS cancer between 2010 and 2018 were identified from the Ontario Cancer Registry. IMD was identified by scanning hospital administrative databases for cranial irradiation or coding for a secondary brain malignancy (ICD-10 code C793).

RESULTS

25,478 of 601,678 (4.2%) patients with a diagnosis of primary cancer were found to have IMD. The median time from primary cancer diagnosis to IMD was 5.2 (0.7, 15.4) months and varied across disease sites, for example, 2.1 months for lung, 7.3 months for kidney, and 22.8 months for breast. Median survival following diagnosis with IMD was 3.7 months. Lung cancer accounted for 60% of all brain metastases, followed by breast cancer (11%) and melanoma (6%). More advanced stage at diagnosis and younger age were associated with a higher likelihood of developing IMD (P < .0001). IMD was also associated with triple-negative breast cancers and ductal histology (P < .001), and with small-cell histology in patients with lung cancer (P < .0001). The annual incidence of IMD was 3,520, translating to 24.2 per 100,000 persons.

CONCLUSION

IMD represents a significant burden in patients with systemic cancers and is a significant cause of cancer mortality. Our findings support measures to actively capture incidents of brain metastasis in cancer registries.

The development and validation of a nomogram for predicting brain metastases in lung squamous cell carcinoma patients: an analysis of the Surveillance, Epidemiology, and End Results (SEER) database

Background

The incidence of brain metastasis (BM) in patients suffering from lung squamous cell carcinoma (LUSC) is lower than that in patients suffering from non-squamous cell carcinoma (NSCC) and there are few studies on BM of LUSC. The purpose of this investigation was to ascertain the risk factors of LUSC, as well as to establish a nomogram prognostic model to predict the incidence of BM in patients with LUSC.

Methods

Patients diagnosed with LUSC between 2010 and 2015 were identified from the Surveillance, Epidemiology, and End Results (SEER) database and the patient data were collated. All patients diagnosed from 2010–2012 were allocated into the training cohort, and the remaining patients diagnosed from 2013–2015 formed the test cohort. Using factors that were screened out through logistic regression analyses, the nomogram in the training cohort was established. It was then evaluated for discrimination and calibration using the test cohort. The performance of the nomogram was assessed by quantifying the area under the receiver operating characteristic (ROC) curve and evaluating the calibration curve.

Results

A total of 26,154 LUSC patients were included in the study. The training cohort consisted of 16,543 patients and there were 8611 patients in the test cohort. Age, marital status, insurance status, histological grade, tumor location, laterality, stage of the cancer, number of metastatic organs, chemotherapy, surgery, and radiotherapy were highly correlated with the incidence of BM. The area under the ROC curve (AUC) of the nomogram for the training cohort and the test cohort were 0.810 [95% confidence interval (CI): 0.796 to 0.823] and 0.805 (95% CI: 0.784 to 0.825), respectively. The slope of the calibration curve was close to 1.

Conclusions

The nomogram was able to accurately predict the incidence of BM. This may be beneficial for the early identification of high-risk LUSC patients and the establishment of individualized treatments.