nscl thesis database

Maintenance work is planned for Wednesday 1st May 2024 from 9:00am to 11:00am (BST).

During this time, the performance of our website may be affected - searches may run slowly and some pages may be temporarily unavailable. If this happens, please try refreshing your web browser or try waiting two to three minutes before trying again.

We apologise for any inconvenience this might cause and thank you for your patience.

New Journal of Chemistry

Harvesting 62 zn from an aqueous cocktail at the nscl †.

* Corresponding authors

a Department of Chemistry, Michigan State University, 578 S Shaw Ln, East Lansing, MI 48824, USA E-mail: [email protected] Tel: +1-517-353-1087

b National Superconducting Cyclotron Laboratory, Michigan State University, 640 S Shaw Ln, East Lansing, MI 48824, USA E-mail: [email protected] , [email protected] , [email protected] , [email protected] , [email protected] , [email protected]

“Isotope harvesting” is a technique that offers access to exotic radionuclides created as by-products during nuclear science research. Ongoing exploratory work at the National Superconducting Cyclotron Laboratory (NSCL) is directed towards the production and extraction of rare radionuclides from a flowing-water target and intends to pave the way for future harvesting efforts at the upcoming Facility for Rare Isotope Beams (FRIB). Here we present the collection of 62 Zn from an aqueous matrix irradiated with a 150 MeV per nucleon 78 Kr beam, while synergistically capturing other gaseous reaction products. In addition to the production rate for 62 Zn (9.08(30) × 10 −5 62 Zn per incoming 78 Kr), the rates of formation for several other radionuclides were determined as well. The purification of 62 Zn from a large number of co-produced radionuclides was performed by anion exchange chromatography, allowing the isolation of 80.5(5.2)% of the generated 62 Zn. With the decay of 62 Zn the radioactive daughter 62 Cu is generated, and with the isolation of pure 62 Cu eluate, the principle of a medical radionuclide generator could be demonstrated. To illustrate the applicability of the obtained 62 Zn, the isolated product was used in free and DTPA-labelled form in a proof of principle plant uptake study with garden cress employing phosphor imaging for visualization.

Supplementary files

• Supplementary information PDF (1460K)

Article information

Download citation, author version available, permissions.

Harvesting 62 Zn from an aqueous cocktail at the NSCL

K. A. Domnanich, C. K. Vyas, E. P. Abel, C. Kalman, W. Walker and G. W. Severin, New J. Chem. , 2020,  44 , 20861 DOI: 10.1039/D0NJ04411C

To request permission to reproduce material from this article, please go to the Copyright Clearance Center request page .

If you are an author contributing to an RSC publication, you do not need to request permission provided correct acknowledgement is given.

If you are the author of this article, you do not need to request permission to reproduce figures and diagrams provided correct acknowledgement is given. If you want to reproduce the whole article in a third-party publication (excluding your thesis/dissertation for which permission is not required) please go to the Copyright Clearance Center request page .

Read more about how to correctly acknowledge RSC content .

Social activity

Search articles by author, advertisements.

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• My Account Login
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 06 October 2022

Identidication of novel biomarkers in non-small cell lung cancer using machine learning

• Fangwei Wang 1 ,
• Qisheng Su 2 &
• Chaoqian Li 1  

Scientific Reports volume  12 , Article number:  16693 ( 2022 ) Cite this article

5957 Accesses

10 Citations

2 Altmetric

Metrics details

• Bioinformatics

Lung cancer is one of the leading causes of cancer-related deaths worldwide, and non-small cell lung cancer (NSCLC) accounts for a large proportion of lung cancer cases, with few diagnostic and therapeutic targets currently available for NSCLC. This study aimed to identify specific biomarkers for NSCLC. We obtained three gene-expression profiles from the Gene Expression Omnibus database (GSE18842, GSE21933, and GSE32863) and screened for differentially expressed genes (DEGs) between NSCLC and normal lung tissue. Enrichment analyses were performed using Gene Ontology, Disease Ontology, and the Kyoto Encyclopedia of Genes and Genomes. Machine learning methods were used to identify the optimal diagnostic biomarkers for NSCLC using least absolute shrinkage and selection operator logistic regression, and support vector machine recursive feature elimination. CIBERSORT was used to assess immune cell infiltration in NSCLC and the correlation between biomarkers and immune cells. Finally, using western blot, small interfering RNA, Cholecystokinin-8, and transwell assays, the biological functions of biomarkers with high predictive value were validated. A total of 371 DEGs (165 up-regulated genes and 206 down-regulated genes) were identified, and enrichment analysis revealed that these DEGs might be linked to the development and progression of NSCLC. ABCA8, ADAMTS8, ASPA, CEP55, FHL1, PYCR1, RAMP3 , and TPX2 genes were identified as novel diagnostic biomarkers for NSCLC. Monocytes were the most visible activated immune cells in NSCLC. The knockdown of the TPX2 gene, a biomarker with a high predictive value, inhibited A549 cell proliferation and migration. This study identified eight potential diagnostic biomarkers for NSCLC. Further, the TPX2 gene may be a therapeutic target for NSCLC.

Similar content being viewed by others

High-content CRISPR screening

Christoph Bock, Paul Datlinger, … Xiaowei Zhuang

An atlas of epithelial cell states and plasticity in lung adenocarcinoma

Guangchun Han, Ansam Sinjab, … Humam Kadara

Cell-type-specific and disease-associated expression quantitative trait loci in the human lung

Heini M. Natri, Christina B. Del Azodi, … Nicholas E. Banovich

Introduction

Non-small cell lung cancer (NSCLC), a subtype of lung cancer, is one of the most prevalent malignancies worldwide. According to studies, the prognosis for NSCLC is highly dependent on the stage of disease progression, and the earlier the disease is detected, the better the chances of survival within 5 years 1 . Patients with early-stage lung cancer often have no obvious symptoms and thus miss the best time for treatment. Moreover, metastasis is the most devastating feature of the tumor, ultimately leading to a high mortality rate 2 . Although some advances in lung cancer treatment and pharmaceutical research have been made, serious complications such as anemia and neutropenia persist, and recurrence rates and mortality in NSCLC patients are still not effectively controlled 3 . Therefore, there is an urgent need to identify reliable biomarkers in the diagnosis and prognosis of NSCLC.

With the significant advancement of microarray and sequencing technology in recent years, gene characterization based on messenger RNA expression levels has shown great promise in diagnosing cancer. For instance, the breast cancer susceptibility protein-1 ( BRCA1 ) gene has been considered a predictor in breast cancer risk models. It is used as a clinical genetic test standard, while the tumor germination ( Bd ) gene is also considered a prognostic biomarker and has an independent prognostic value for disease-free survival (DFS) and overall survival (OS) in colon cancer 4 .

It has previously been reported that bioinformatics methods are used to analyze comprehensive gene expression data to obtain cancer-related biomarkers for effective prevention, diagnosis, and treatment of cancer. Machine learning (ML) methods are a branch of bioinformatics applied to various aspects of cancer research and are a hot topic in lung cancer research. For example, ML methods successfully developed and validated a predictive model for cancer-related deep vein thrombosis 5 . Lai et al. used ML methods to create gene signatures that accurately predicted prostate cancer prognosis 6 . Zheng et al. developed an integrated radiomic model using the ML method to predict the prognosis of prostate cancer patients 7 . ML methods have also been used to screen for potential biomarkers of prostate cancer and osteosarcoma 8 , 9 . Support vector machines (SVM), a linear classifier with maximum intervals defined in the feature space, have gradually been applied to machine learning of cancer using its efficient two-class model to classify cancer with high accuracy from a large number of genomes and efficiently extract key genes to achieve high-accuracy, low-error sample classification. In recent years, many researchers have used SVM methods to study cancer and have made significant advances. For example, Luo et al. used SVM methods to construct predictive models for synchronous lung metastases (SLM) in osteosarcoma 10 . Su et al. identified eight genes associated with colon cancer prognosis using the SVM method 11 . Cai et al. used SVM methods to construct radiomics-based models for diagnosing lung adenocarcinoma (LUAD) 12 . SVM methods also differentiated between epithelial ovarian cancer (EOC) and surrounding tissues, aiding EOC diagnosis 13 . In addition, the SVM methods successfully screened colorectal biomarkers for personalized treatment of patients with post-operative liver metastases 14 . However, there have been few studies on potential NSCLC biomarkers.

In this study, we obtained differentially expressed genes (DEGs) for NSCLC from the Gene Expression Omnibus (GEO) database. We used functional enrichment analysis to identify NSCLC-related biomarkers using an ML approach, followed by infiltrative immune cell analysis and in vitro validation of the biomarkers’ functions. Our findings may be useful in the early diagnosis of NSCLC and the mechanistic study of NSCLC.

Materials and methods

Data processing and degs screening.

We obtained three microarray datasets (GSE18842, GSE32863, and GSE21933) from the Gene Expression Omnibus (GEO) public databases, and the organism parameter was set to “Homo sapiens”. The raw data from these datasets were processed using R language statistical software (version 4.1.2) 15 . Using the R ‘Limma’ package 16 , we identified DEGs between normal lung and NSCLC tissues, with an adjusted P -value < 0.05 and |logFC|≥ 2 as statistical significance. Heatmaps and volcano plots were plotted to show the differential expression of DEGs.

Functional and pathways enrichment analysis

Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG) 17 , 18 , 19 , and Disease Ontology (DO) enrichment analyses of DEGs were performed using the R ‘clusterProfiler’ package 20 and visualized using the R ‘ggplot2’ package 21 with GO functional annotation, including biological process, cellular component, and molecular function terms. The R ‘clusterProfiler’ and R ‘org.Hs.eg.db’ packages 22 were used to enrich DEGs using gene set enrichment analysis (GSEA). A result with an adjusted P -value < 0.05 and a false discovery rate < 0.05 was considered statistically different.

ML methods to identify NSCLC biomarkers

For biomarker screening, two ML methods were used: least absolute shrinkage and selection operator (LASSO) logistic regression 23 and support vector machine-recursive feature elimination (SVM-RFE) 24 . The algorithm LASSO used the R ‘glmnet’ package 25 , while the SVM-RFE algorithm used the R ‘e1071’ package 26 . The following are the model settings: LASSOcvfit = cv.glmnet (x,y,family = ‘binomial,” alpha = 1, type.measure = ‘deviance,’ nfolds = 10). SVM = rfeControl (functions = caretFuncs, method = “cv,” methods = “svmRadial”). The point with the lowest cross-validation in the vertical axis corresponds to the biomarker genes to be found; a difference of P  < 0.05 was considered statistically significant.

Validation of biomarkers

The R ‘ggpubr’ package 27 was used to examine the biomarker expression in GSE32864. In addition, the biomarkers’ diagnostic efficiency was validated using receiver operating characteristic (ROC) curves generated with the R ‘pROC’ package 28 ; P  < 0.05 was considered statistically significant.

Assessment and correlation analysis of infiltrating immune cells

The CIBERSORT algorithm 29 was used to analyze the relationship between infiltrating immune cells and biomarkers; a correlation heatmap was produced using the R ‘corrplot’ package 30 to detect the association of each immune cell with the other cells in the LUAD sample; the violin map using the R “ggplot2” package showed differences in the expression of 22 immune cells. A Spearman correlation analysis was performed between diagnostic biomarkers and infiltrating immune cells using the R “ggstatsplot” package.

Survival analysis

The GEIPA online website ( http://gepia.cancer-pku.cn/ ) is a dataset based on The Cancer Genome Atlas and Genotype-Tissue Expression that provides a fast and customizable web-based tool. A summary of the process is as follows: click on “survival plots” and enter the gene name, then select “LUAD,” “OS,” and “RFS,” and finally observe the P -value and output the graph; P  < 0.05 was considered statistically significant.

Cell culture and cell transfection

A549 cells were purchased from the ATCC (Shanghai, China). A549 cells were cultured in Dulbecco’s modified Eagle medium (Thermo Fisher Scientific, USA) containing 10% serum (Thermo Fisher Scientific, Wilmington, DE, USA). For si-RNA transfection, A549 cells were transfected with si-TPX2 using the Lipofectamine 2000 transfection reagent (Invitrogen, Waltham, MA, USA) according to the manufacturer’s instructions. To target TPX2, the following si-RNA sequences were used: AGCCTCAGAAGATCTCTTAG (si-TPX2).

Western blotting

The total protein concentration extracted with lysis buffer containing protease inhibitors was measured using the bicinchoninic acid (BCA) protein assay kit (Beyotime Biotechnology Inc., Shanghai, China), and proteins were separated using a polyvinylidene fluoride membrane (Millipore, Billerica, MA, USA). Proteins were separated on 10% skim sodium dodecyl sulfate–polyacrylamide gel electrophoresis, blocked with 20% skim milk and incubated with primary antibody overnight at 4 °C. Western blotting was performed using an enhanced chemiluminescence detection reagent (Beyotime Biotechnology Inc., Shanghai, China.) and according to the manufacturer’s protocol. P  < 0.05 was considered statistically significant.

Statistical analysis

All statistical analyses were performed using the SPSS version 21.0 software package (SPSS, Chicago, Il, USA). The data are expressed as mean ± standard deviation. Categorical variables were analyzed using the χ2 or Fisher’s exact test. For paired samples, continuous variables were analyzed using the student’s t-test, and differences between groups were analyzed using an analysis of variance calculation. When the basic assumptions of the student’s t-test were not satisfied, the Wilcoxon–Mann–Whitney test was used. P -value < 0.05 was considered to indicate a statistically significant difference.

Identification of DEGs

To analyze the diagnostic genes of NSCLC, we designed a flowchart (Fig.  1 ). Using the ‘Limma’ package in the R language, 371 DEGs, including 165 up-regulated genes and 206 down-regulated genes, were screened in NSCLC and normal lung tissue samples of GSE18842, GSE32864, and GSE21933 according to the criteria (adjusted P -value < 0.05 and |logFC|≥ 1). The results were expressed in the heatmap (Fig.  2 a) and a volcano plot (Fig.  2 b).

a flowchart of the entire analysis process to the manuscript.

Identification of DEGs. ( a ) Heatmap of DEGs. ( b ) Volcano diagram of DEGs, red indicates up-regulated, blue indicates down-regulated.

Functional enrichment analyses of the DEGs

We investigated the possible biological functions of these 371 DEGs using the GO, KEGG, DO, and GSEA functional enrichment analyses. The GO analysis revealed that DEGs were primarily enriched in nuclear division and extracellular matrix organization, implying a link between tumor cell division and distant tumor metastasis (Fig.  3 a). According to KEGG pathway analysis, the IL-17 signaling pathway, cell cycle, complement and coagulation cascades, and malaria were four significantly enriched pathways (Fig.  3 b). The DO analysis revealed that these DEGs were remarkably enriched in lung disease, NSCLC, and integumentary system disease (Fig.  3 c). Finally, the GSEA analysis revealed that in lung cancer tissue, complement and coagulation cascades, cytokine-cytokine receptor interactions, hematopoietic cell lineage, leukocyte transendothelial migration, lysosome, and natural killer cell-mediated cytotoxicity were highly active (Fig.  3 d), whereas base excision repair, cell cycle, DNA replication, mismatch repair, p53 signaling pathway, and pyrimidine metabolism were highly active in normal nasopharyngeal tissue (Fig.  3 e). All these findings indicate that these DEGs may be critical in NSCLC.

Functional enrichment analysis of DEGs. The P value represents the colour depth of the node. The size of the node implies the number of DEGs. ( a ) GO functional enrichment analysis results for DEGs, including Biological process(BP), molecular function(MF) and cellular component(CC). ( b ) KEEG enrichment analysis reveals signalling pathways highly relevant to NSCLC. ( c ) DO functional enrichment analysis results of DEGs. ( d ) GSEA shows the top six signalling pathways most associated with normal lung tissue. ( e ) GSEA shows the top six signalling pathways most associated with NSCLC.

Screening for NSCLC biomarkers and validation

The LASSO logistic algorithm was used in this study to identify, 26 characteristic genes (Fig.  4 a), while the SVM-RFE method was used to identify 40 characteristic genes (Fig.  4 b). Eight biomarker genes were obtained as a result of the intersection, ABCA8 , ADAMTS8 , ASPA , CEP55 , FHL1 , PYCR1 , RAMP3 , and TPX2 genes (Fig.  4 c). To further validate their potential as diagnostic biomarkers for NSCLC, we examined their expression in the GSE32863 dataset (Fig.  5 ), which revealed that ABCA8 , ADAMTS8 , ASPA , FHL1 , and RAMP3 genes were down-regulated in NSCLC while CEP55 , PYCR1 , and TPX2 genes were up-regulated. The accuracy of these eight biomarkers in distinguishing NSCLC from normal individuals was evaluated using a receiver operating characteristic (ROC) analysis, and all eight biomarkers demonstrated high sensitivity and specificity. (The areas under the ROC curves (AUCs) = 0.999 in GSE18842 and GSE 21993 and 0.910 in GSE32863 for the ABCA8 gene; AUCs = 0.998 in GSE18842 and GSE21993 and 0.930 in GSE32863 for the ADAMTS8 gene, AUCs = 0.996 in GSE18842 and GSE 21993 and 0.941 in GSE32863 for the ASPA gene, AUCs = 0.998 in GSE18842 and GSE21993 and 0.904 in GSE32863 for the CEP55 gene, AUCs = 0.998 in GSE18842 and GSE21993 and 0.945 in GSE32863 for the FHL1 gene, AUCs = 0.998 in GSE18842 and GSE21993 and 0.921 in GSE32863 for the PYCR1 gene, AUCs = 0.993 in GSE18842 and GSE21993 and 0.936 in GSE32863 for the RAMP3 gene, AUCs = 0.998 in GSE18842 and GSE21993 and 0.895 in GSE32863 for the TPX2 gene. All P  < 0.05) (Figs.  6 and 7 ).

Machine learning approach to screen for NSCLC-related biomarkers. The point corresponding to the smallest vertical coordinate is the characteristic genes. ( a ) Results of screening biomarkers based on LASSO algorithm. ( b ) Screening results for biomarkers based on the SVM-RFE algorithm. ( c ) The Venn diagram shows the results of the intersection of the LASSO algorithm and the SVM-RFE algorithm, with the intersection resulting in eight biomarkers.

Expression of eight biomarkers in the validation group (GSE32863). ( a ) ABCA8. ( b ) ADAMTS8. ( c ) ASPA. ( d ) CEP55. ( e ) FHL1. ( f ) PYCR1. ( g ) RAMP3. ( h ) TPX2. P  < 0.05 means difference is statistically significant.

ROC curves for eight biomarkers in the training dataset(GSE18842 and GSE21933). ( a ) ABCA8. ( b ) ADAMTS8. ( c ) ASPA. ( d ) CEP55. ( e ) FHL1. ( f ) PYCR1. ( g ) RAMP3. ( h ) TPX2.

ROC curves for eight biomarkers in the validation dataset (GSE32863). ( a ) ABCA8. ( b ) ADAMTS8. ( c ) ASPA. ( d ) CEP55. ( e ) FHL1. ( f ) PYCR1. ( g ) RAMP3. ( h ) TPX2.

Assessment of immune cell infiltration

A comprehensive and dynamic understanding of the immune microenvironment is essential to develop effective therapeutic strategies. Therefore, in this study, we investigated immune cell infiltration in NSCLC and the relationship between biomarkers and infiltrating immune cells. First, we found significant differences in the composition of the 22 infiltrating immune cell types in each tissue sample (Fig.  8 a). The correlation matrix showed the strongest positive correlation between eosinophils and monocytes and the strongest negative correlation between macrophage M0 cells and monocytes (Fig.  8 b). Monocytes and eosinophils were the most down-regulated cells in NSCLC, while plasma cells and macrophage M0 were the most up-regulated, and activation of neutrophils, macrophage M1, and NK cells was low (Fig.  8 c). Figure  8 depicts the relationship between the expression of eight biomarkers and the infiltration of immune cells. Monocytes, eosinophils, NK cells activated, neutrophils, mast cells resting, and T cells CD4 memory resting were positively correlated with with the ABCA8 gene expression, whereas plasma cells, macrophages M0, Follicular helper T (Tfh) cells, and macrophages M1 were negatively correlated (Fig.  9 ). The ADAMTS8 expression levels were negatively correlated with macrophages M1 T cells follicular helper, macrophages M0, and plasma cells, and positively correlated with monocytes, eosinophils, NK cells activated, neutrophils, T cells CD8, and T cells CD4 memory resting. The with the APSA gene expression levels expression levels were negatively correlated with T cells follicular helper, macrophages M0, and plasma cells and positively correlated with monocytes, eosinophils, neutrophils, mast cells resting, and NK cells activated. With The CEP55 gene expression levels expression levels were positively correlated with plasma cells, macrophages M0, macrophages M1, and T cells follicular helper and negatively correlated with T cells CD8, NK cells activated, mast cells resting, neutrophils, eosinophils, and monocytes. With The FHL1 gene expression levels expression levels were positively correlated with monocytes, eosinophils, NK cells activated, neutrophils, T cells CD4 memory resting, mast cells resting, and T cells CD8 and negatively correlated with T cells gamma delta, T cells regulatory, T cells CD4 memory activated, T cells follicular helper, macrophages M0, and plasma cells. With the PYCR1 gene expression levels were positively correlated with plasma cells, macrophages M0, T cells regulatory, T cells follicular helper, and macrophages M1 and negatively correlated with mast cells resting, T cells CD4 memory resting, NK cells activated, neutrophils, eosinophils, and monocytes. With The RAMP3 gene expression levels were positively correlated with monocytes, eosinophils, mast cells resting, T cells CD8, NK cells activated, and neutrophils, and negatively correlated with T cells follicular helper, macrophages M0, and plasma cells. The TPX2 expression levels were positively correlated with plasma cells, macrophages M0, macrophages M1, T cells follicular helper, and B cells naive and negatively correlated with T cells CD4 memory resting, NK cells activated, mast cells resting, neutrophils, eosinophils, and monocytes ( P  < 0.01).

Correlation analysis among immune cells and Differential analysis of immune cells. ( a ) The 22 immune cell populations in normal lung tissue and NSCLC. ‘con’ represents normal lung tissue, ‘treat’ represents NSCLC. The vertical coordinate represents the amount of immune cells. ( b ) Correlation heatmap Indicates the correlation analysis of immune cells. The redder of the point, the stronger the positive correlation, the reddest point represents the two immune cells with the most significant positive correlation. The bluer the colour of the point, the stronger the negative correlation, the two immune cells with the most significant negative correlation correspond to the bluest points. ( c ) The vioplot showed the difference in 22 immune cells between normal lung tissue and NSCLC. ‘con’ represents normal lung tissue, ‘treat’ represents NSCLC. P  < 0.05 means difference is statistically significant.

Correlation analysis between eight biomarkers and immune cells. Colours represent P -values, P  < 0.05 means significant correlation and is indicated in red, the size of the circle represents the absolute value of the correlation coefficient. ( a ) ABCA8. ( b ) ADAMTS8. ( c ) ASPA. ( d ) CEP55. ( e ) FHL1. ( f ) PYCR1. ( g ) RAMP3. ( h ) TPX2.

To determine the potential prognostic value of these eight biomarkers, we investigated the relationship between each biomarker’s expression and OS and DFS in NSCLC. The findings revealed that ABCA8 and FHL1 genes were associated with longer OS in lung cancer patients ( P  < 0.05), whereas TPX2 and CEP55 genes were associated with shorter OS ( P  < 0.05) (Figs.  10 a–d), and other biomarkers were not significantly associated with OS ( P  > 0.05). Notably, only the TPX2 gene was associated with poor DFS (Fig.  10 e), suggesting that the TPX2 gene may have a potential prognostic value in NSCLC, which we investigated further.

Survival analysis of eight biomarkers in GEPIA website. ( a–d ) Overall survival analysis of FLHI, ABCA8, CEP55, TPX2. ( e ) Disease Free survival analysis of TPX2. P  < 0.05 means difference is statistically significant.

Knockdown of the TPX2 gene inhibited the proliferation and migration of A549 cells

To investigate the potential role of the TPX2 gene in NSCLC, we used a western blot to compare the expression of TPX2 in A549 cells and BEAS-2B cells. We found that A549 cells had higher levels of TPX2 protein expression than BEAS-2B cells ( P  < 0.05) (Fig.  11 a, b). Furthermore, we inhibited TPX2 expression by transfecting si-RNA-targeted TPX2 into A549 cells. The results of western blot analysis revealed that the levels of TPX2 proteins were significantly lower in A549 cells after transfection with si-TPX2 compared to si-control ( P  < 0.05) (Fig.  11 c, d), indicating that the transfection was complete and ready for the next step of the experiment. The CCK8 proliferation assays confirmed that the TPX2 gene knockdown significantly inhibited A549 cell proliferation ( P  < 0.05) (Fig.  12 ). Moreover, the results of the transwell migration assay revealed that the relative number ratios of migrating cells were significantly lower in the TPX2 gene knockdown cells compared to si-control cells ( P  < 0.05) (Fig.  13 ), indicating that the TPX2 gene knockdown resulted in A549 cell migration ability. Therefore, we speculated that the TPX2 gene might act as an oncogene, promoting NSCLC progression. However, this conclusion needs to be verified via in vivo experiments.

( a , b ) Western blot method to detect TPX2 protein expression in normal alveolar epithelial cells BEAS-2B and non-small cell lung cancer cells A549. ( c , d ) Western blot method to detect si-TPX2 transfection efficiency. ‘*’represents P  < 0.05.

CCK8 method to detect the proliferative ability of the si-NC and si-TPX2 groups. Silencing of TPX2 inhibited the proliferative capacity of A549 cells. ‘*’ represents P  < 0.05.

Transwell method to detect migration ability of si-NC group and si-TPX2 group. Silencing of TPX2 suppressed migration ability in A549 cells. ‘*’ represents P  < 0.05.

NSCLC is well known for being asymptomatic and can only be detected at an early stage through physical examination 1 . As the tumor grows, develops, and spreads, serious symptoms emerge, including chest pain, breathing difficulties, liver metastases, and a slew of seriously life-threatening symptoms. Given the lack of obvious symptoms in the early stages of NSCLC, which makes diagnosis difficult, “early diagnosis and early treatment” has become the treatment consensus for NSCLC 31 . With the advancement of bioinformatics, effective analysis and exploration of cancer genes to find tumor biomarkers have become a hot topic for early cancer diagnosis expression profiles and treatment, For example, chen et al. have used new computational models in the field of miRNA to make great contributions to the pathogenesis of diseases and new drug development. For example, through the construction of Neighborhood Constraint Matrix Completion for MiRNA-Disease Association prediction (NCMCMDA), deep-belief network for miRNA-disease association prediction (DBNMDA), Ensemble of Decision Tree based MiRNA-Disease Association prediction (EDTMDA) models to accurately predict the potential relationship between miRNA-disease and in breast neoplasms, lung neoplasms, esophageal neoplasms have been validated 32 , 33 , 34 . The study of chen et al. has greatly improved the experimental efficiency and provided a new theoretical basis for the prevention, diagnosis and treatment of complex human diseases by screening disease-associated miRNAs through computational models, but such methods have not been adequately studied and described for biomarkers of NSCLC progression. Therefore, it is critical to identify a sensitive, safe, and feasible NSCLC biomarker for diagnostic and therapeutic purposes and to improve patient survival 35 .

The bioinformatics analysis of the microarray dataset from the GEO database identified 165 up-regulated and 206 down-regulated genes between NSCLC and normal lung tissue samples. Moreover, functional analyses revealed that these DEGs were linked to lung cancer tumorigenesis and metastasis. The most significantly enriched pathway, the cell cycle signaling pathway, has been shown to have mutations that can affect the genomic and microenvironmental characteristics of LUAD patients 36 and can also be used as an assessment criterion for post-operative adjunctive therapy in LUAD patients 37 . Furthermore, GSEA analysis revealed that these DEGs might affect base excision repair, the cell cycle, DNA replication, mismatch repair, and the p53 signaling pathway. The most activated p53 signaling pathway, which has been linked to oncogenic effects 38 , promotes tumor growth in NSCLC and pancreatic ductal adenocarcinoma (PDAC) 39 , 40 . However, these DEGs’ specific functions and molecular mechanisms need to be investigated further.

Because of its flexibility and power, ML is increasingly being used to screen novel biomarkers. We used two recently popular ML approaches, LASSO logistic regressions, and SVM-RFM, to identify the best diagnostic biomarkers for NSCLC. After combining the two methods and ROC analysis, eight NSCLC-related biomarkers with accurate predictive properties were identified (the AUCs of all these eight genes were greater than 0.89), including ADAMTS8 , ABCA8 , TPX2 , CEP55 , ASPA , FHL1 , RAMP3 , and PYCR1 genes. A disintegrin and metallopeptidase with thrombospondin motif type 8 (ADAMTS8) is a member of the zinc metalloproteinase family and is considered a tumor suppressor 41 . Wu et al. found that ADAMTS8 has been associated with clinical staging and lymph node metastasis in esophageal cancer patients 42 . ADAMTS8 can also inhibit lung cancer by targeting Vascular endothelial growth factor (VEGFA) 43 . ATP-binding cassette subfamily A 8 (ABCA8) has been linked to tumors. Studies have shown that ABCA8 can inhibit the proliferation of breast cancer cells by regulating the AMPK/mTOR signaling pathway 44 , and ABCA8 can also be used as a prognostic marker for hepatocellular carcinoma 45 and gastric adenocarcinoma 46 . Centrosomal protein, 55 kD (CEP55) has been reported to be a potential biomarker and therapeutic target for PDAC and lung cancer 47 . Several studies have shown that CEP55 is carcinogenic in the colon and esophageal cancers 48 , 49 . Targeting protein for xenopus kinesin-like protein 2 (TPX2) is a cell cycle-associated gene that plays a pro-oncogenic role in hepatocellular carcinoma cells and acts synergistically with anti-cancer drugs 50 . TPX2 can also be a therapeutic target for breast cancer and is associated with patient prognosis 51 . Current studies have shown that recombinant Helicobacter Pylori aspartate ammonia-lyase (ASPA) is an effective predictor of prognosis in colorectal cancer patients 52 . Four-and-a-half LIM domains protein (FHL1) has a diagnostic value in microscopic papillary thyroid carcinoma 53 , and it has been reported that FHL1 acts as an inhibitor in colorectal cancer cells 54 , 55 , 56 . Receptor activity-modifying protein 3 (RAMP3) is an accessory molecule that forms complexes with and regulates the function of specific G protein-coupled receptors (GPCRs). To date, studies have shown that RAMP3 is overexpressed in hepatocellular carcinoma patients and that RAMP3 is an independent prognostic factor for overall survival and RFS 57 . Pyrroline-5-Carboxylate Reductase 1 (PYCR1) is a mitochondrial enzyme that is the final step in the proline biosynthetic pathway. Currently, PYCR1 has been reported to regulate tumour cell proliferation in hepatocellular carcinoma and cloud be an effective therapeutic target for multiple myeloma 58 , 59 , as well as inhibit the development of clear cell renal cell carcinoma by causing mitochondrial dysfunction and interfering with oxidative stress pathways 60 . Thus, according to the available studies, the eight biomarkers mentioned above play an important role in tumorigenesis and progression, but their exact mechanisms in NSCLC remain unknown.

Given the importance of the immune microenvironment in the development of lung cancer, we also performed immunological analyses. We found that monocytes differed most between normal and NSCLC tissue samples and correlated closely with all eight previously obtained biomarkers, implying that monocytes may be the most active immune cells in NSCLC. Monocytes have been shown to play a role in tumor progression 61 . For example, up-regulation of monocytes promotes CT26 tumor progression 62 , 63 . It has also been linked to the immune checkpoint matrix metalloproteinases in hepatocellular carcinoma 64 . The findings of this study suggest that monocytes may contribute to clinical immunotherapy and warrant further investigation.

Another significant finding in our study was that TPX2 expression was linked to poor prognosis, and TPX2 was up-regulated in both bioinformatics and western blot validation. We also found that TPX2 promoted the proliferation and migration of lung cancer A549 cells in vitro, suggesting that TPX2 was involved in developing NSCLC and could be a therapeutic target for NSCLC. However, the exact mechanism remains unknown, and we will investigate the role of TPX2 in vivo experiments.

ML methods have been successfully applied in cancer research in recent years, for example, for cancer classification 65 , for analyzing gene chip data to screen for cancer-related biomarkers to assist doctors in making better diagnoses and decisions 66 , and for cancer prediction to significantly improve prediction accuracy 65 . Notably, the ML method has been demonstrated to be effective in determining a non-invasive, accurate, and reliable diagnosis of NSCLC. Li et al. used specific carbonyl volatile organic compounds in exhaled breath as a biomarker for detecting lung cancer to distinguish lung cancer patients from healthy controls and patients with benign lung nodules 66 . Zhang et al. suggest that 5-hydroxymethylcytosine in circulating cell-free DNA can be used to diagnose and treat NSCLC 67 . Zhang et al. concluded that five circulating micro RNAs have important prognostic capabilities in lung cancer 68 . Wang et al. identified eight differentially expressed long non-coding RNAs in LUAD that could be used as potential diagnostic biomarkers 69 .

This study combines machine learning methods (LASSO algorithm and SVM-RFE algorithm) with medical experiments to identify non-small cell lung cancer (NSCLC)-related biomarkers through the GEO public database, and finds that the progression of NSCLC is associated with immune cell infiltration, which greatly improves the efficiency of basic experiments for clinicians, and the screened NSCLC-related biomarkers is important for the prevention, diagnosis and treatment of NSCLC. Subsequently, we used in vitro functional assays to silence the prognostic TPX2 biomarker gene in NSCLC A549 cells and found that the proliferation and migration ability of A549 cells were significantly reduced, which further confirmed that the NSCLC-related biomarkers screened by the machine learning approach are novel and reliable. Our study provides a new idea for the pathogenesis and targeted therapy of NSCLC.

In conclusion, we used machine learning methods to identify eight diagnostic biomarkers for NSCLC, including ADAMTS8 , ABCA8 , TPX2 , CEP55 , ASPA , FHL1 , RAMP3 , and PYCR1 genes, followed by functional enrichment analysis and immune correlation analysis, and we validated the potential role of the TPX2 gene in vitro. Our findings identify new potential biomarkers for diagnosing and treating NSCLC and reveal new approaches that may have therapeutic potential for NSCLC.

Data availability

The datasets generated during the current study are available in the Gene Expression Omnibus (GEO) repository, ( https://www.ncbi.nlm.nih.gov/geo/ ), ( https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE18842 ), ( https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE21933 ), ( https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE32863 ).

Meador, C. B. & Lovly, C. M. A tale of two histologies: Dissecting the biology of lineage transformation in lung cancer. Cancer Discov. 11 , 2962–2964 (2021).

Article   PubMed   PubMed Central   Google Scholar  

Esfahani, M. S. et al. Inferring gene expression from cell-free DNA fragmentation profiles. Nat. Biotechnol. https://doi.org/10.1038/s41587-022-01222-4 (2022).

Dai, J. et al. Sleeve resection after neoadjuvant chemoimmunotherapy in the treatment of locally advanced non-small cell lung cancer. Transl. Lung Cancer Res. 11 , 188–200 (2022).

Article   CAS   PubMed   PubMed Central   Google Scholar  

Basile, D. et al. Tumor budding is an independent prognostic factor in stage III colon cancer patients: A post-hoc analysis of the IDEA-France phase III trial (PRODIGE-GERCOR). Ann. Oncol. https://doi.org/10.1016/j.annonc.2022.03.002 (2022).

Article   PubMed   Google Scholar  

Jin, S. et al. Machine learning predicts cancer-associated deep vein thrombosis using clinically available variables. Int. J. Med. Inform. 161 , 104733 (2022).

Lai, Y.-L. et al. Identification of a steroid hormone-associated gene signature predicting the prognosis of prostate cancer through an integrative bioinformatics analysis. Cancers 14 , 1565 (2022).

Zheng, H. et al. Multiparametric MRI-based radiomics model to predict pelvic lymph node invasion for patients with prostate cancer. Eur. Radiol. https://doi.org/10.1007/s00330-022-08625-6 (2022).

Ayyad, S. M. et al. A new framework for precise identification of prostatic adenocarcinoma. Sensors 22 , 1848 (2022).

Article   ADS   PubMed   PubMed Central   Google Scholar  

Ding, F.-P., Tian, J.-Y., Wu, J., Han, D.-F. & Zhao, D. Identification of key genes as predictive biomarkers for osteosarcoma metastasis using translational bioinformatics. Cancer Cell. Int. 21 , 640 (2021).

Luo, Z. et al. Radiomics analysis of multiparametric MRI for prediction of synchronous lung metastases in osteosarcoma. Front. Oncol. 12 , 802234 (2022).

Su, Y. et al. Colon cancer diagnosis and staging classification based on machine learning and bioinformatics analysis. Comput. Biol. Med. 145 , 105409 (2022).

Cai, J. et al. A radiomics study to predict invasive pulmonary adenocarcinoma appearing as pure ground-glass nodules. Clin. Radiol. 76 , 143–151 (2021).

Article   CAS   PubMed   Google Scholar  

van Vliet-Pérez, S. M. et al. Hyperspectral imaging for tissue classification after advanced stage ovarian cancer surgery-a pilot study. Cancers 14 , 1422 (2022).

Granata, V. et al. EOB-MR based radiomics analysis to assess clinical outcomes following liver resection in colorectal liver metastases. Cancers 14 , 1239 (2022).

R Core Team. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing , (Vienna, Austria, 2017). https://www.R-project.org .

Ritchie, M. E. et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 43 , e47 (2015).

Kanehisa, M. & Goto, S. KEGG: Kyoto encyclopedia of genes and genomes. Nucleic Acids Res. 28 , 27–30 (2000).

Kanehisa, M. Toward understanding the origin and evolution of cellular organisms. Protein Sci. 28 , 1947–1951 (2019).

Kanehisa, M., Furumichi, M., Sato, Y., Ishiguro-Watanabe, M. & Tanabe, M. KEGG: Integrating viruses and cellular organisms. Nucleic Acids Res. 49 , D545–D551 (2021).

Yu, G., Wang, L.-G., Han, Y. & He, Q.-Y. clusterProfiler: An R package for comparing biological themes among gene clusters. OMICS 16 , 284–287 (2012).

Perez-Iratxeta, C., Bork, P. & Andrade-Navarro, M. A. Update of the G2D tool for prioritization of gene candidates to inherited diseases. Nucleic Acids Res. 35 , W212-216 (2007).

Subramanian, A. et al. Gene set enrichment analysis: A knowledge-based approach for interpreting genome-wide expression profiles. Proc. Natl. Acad. Sci. USA 102 , 15545–15550 (2005).

Article   ADS   CAS   PubMed   PubMed Central   Google Scholar  

Tibshirani, R. The lasso method for variable selection in the Cox model. Stat. Med. 16 , 385–395 (1997).

Lin, X. et al. A support vector machine-recursive feature elimination feature selection method based on artificial contrast variables and mutual information. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 910 , 149–155 (2012).

Friedman, J., Hastie, T. & Tibshirani, R. Regularization paths for generalized linear models via coordinate descent. J. Stat. Softw. 33 , 1–22 (2010).

Huang, M.-L., Hung, Y.-H., Lee, W. M., Li, R. K. & Jiang, B.-R. SVM-RFE based feature selection and Taguchi parameters optimization for multiclass SVM classifier. Sci. World J. 2014 , 795624 (2014).

Article   Google Scholar  

Pathan, M. et al. FunRich: An open access standalone functional enrichment and interaction network analysis tool. Proteomics 15 , 2597–2601 (2015).

Tang, X., Zhang, S., Wang, Z., Liu, J. & Ying, Z. ProcData: An R package for process data analysis. Psychometrika 86 , 1058–1083 (2021).

Article   MathSciNet   PubMed   MATH   Google Scholar  

Xue, G., Hua, L., Zhou, N. & Li, J. Characteristics of immune cell infiltration and associated diagnostic biomarkers in ulcerative colitis: Results from bioinformatics analysis. Bioengineered 12 , 252–265 (2021).

Serang, S., Jacobucci, R., Brimhall, K. C. & Grimm, K. J. Exploratory mediation analysis via regularization. Struct. Equ. Modeling 24 , 733–744 (2017).

Article   MathSciNet   PubMed   PubMed Central   Google Scholar  

Oudkerk, M., Liu, S., Heuvelmans, M. A., Walter, J. E. & Field, J. K. Lung cancer LDCT screening and mortality reduction–evidence, pitfalls and future perspectives. Nat. Rev. Clin. Oncol. 18 , 135–151 (2021).

Chen, X., Sun, L.-G. & Zhao, Y. NCMCMDA: miRNA-disease association prediction through neighborhood constraint matrix completion. Brief. Bioinform. 22 , 485–496 (2021).

Chen, X., Li, T.-H., Zhao, Y., Wang, C.-C. & Zhu, C.-C. Deep-belief network for predicting potential miRNA-disease associations. Brief. Bioinform. 22 , bbaa186 (2021).

Chen, X., Zhu, C.-C. & Yin, J. Ensemble of decision tree reveals potential miRNA-disease associations. PLoS Comput. Biol. 15 , e1007209 (2019).

Li, N. et al. One-off low-dose CT for lung cancer screening in China: A multicentre, population-based, prospective cohort study. Lancet. Respir. Med. 10 , 378–391 (2022).

Shan, G. et al. Genomic and tumor microenvironment differences between cell cycle progression pathway altered/non-altered patients with lung adenocarcinoma. Front. Oncol. 12 , 843528 (2022).

Li, J. et al. Identifying 18F-FDG PET-metabolic radiomic signature for lung adenocarcinoma prognosis via the leveraging of prognostic transcriptomic module. Quant. Imaging Med. Surg. 12 , 1893–1908 (2022).

Su, R. et al. A pan-cancer analysis of the oncogenic role of Holliday junction recognition protein in human tumors. Open Med. (Wars) 17 , 317–328 (2022).

Article   CAS   Google Scholar  

Xiao, X. et al. Green tea-derived theabrownin suppresses human non-small cell lung carcinoma in xenograft model through activation of not only p53 signaling but also MAPK/JNK signaling pathway. J. Ethnopharmacol. 291 , 115167 (2022).

Ma, Z. et al. ZMAT1 acts as a tumor suppressor in pancreatic ductal adenocarcinoma by inducing SIRT3/p53 signaling pathway. J. Exp. Clin. Cancer Res. 41 , 130 (2022).

Zhang, K. et al. ADAMTS8 inhibits cell proliferation and invasion, and induces apoptosis in breast cancer. Onco. Targets Ther. 13 , 8373–8382 (2020).

Wu, Z. et al. ADAMTS8 inhibits progression of esophageal squamous cell carcinoma. DNA Cell. Biol. https://doi.org/10.1089/dna.2020.6053 (2020).

Zhang, Y., Hu, K., Qu, Z., Xie, Z. & Tian, F. ADAMTS8 inhibited lung cancer progression through suppressing VEGFA. Biochem. Biophys. Res. Commun. 598 , 1–8 (2022).

Lv, C., Yang, H., Yu, J. & Dai, X. ABCA8 inhibits breast cancer cell proliferation by regulating the AMP activated protein kinase/mammalian target of rapamycin signaling pathway. Environ. Toxicol. https://doi.org/10.1002/tox.23495 (2022).

Zhang, J., Zhang, X., Li, J. & Song, Z. Systematic analysis of the ABC transporter family in hepatocellular carcinoma reveals the importance of ABCB6 in regulating ferroptosis. Life Sci. 257 , 118131 (2020).

Guo, Y., Wang, Z. W., Su, W. H., Chen, J. & Wang, Y. L. Prognostic value and immune infiltrates of ABCA8 and FABP4 in stomach adenocarcinoma. Biomed. Res. Int. 2020 , 4145164 (2020).

Wang, C. et al. Identification of hub genes in pancreatic ductal adenocarcinoma using bioinformatics analysis. Iran J. Public Health 50 , 2238–2245 (2021).

PubMed   PubMed Central   Google Scholar  

Lin, Y., Chen, Y., Shen, R., Chen, D. & Lin, Y. MicroRNA-148a-3p suppresses cell proliferation and migration of esophageal carcinoma by targeting CEP55. Cell. Mol. Biol. Lett. 26 , 54 (2021).

Bozic, D. et al. Predicting sulforaphane-induced adverse effects in colon cancer patients via in silico investigation. Biomed. Pharmacother. 146 , 112598 (2022).

Wang, X., Wang, J., Shen, H., Luo, Z. & Lu, X. Downregulation of TPX2 impairs the antitumor activity of CD8+ T cells in hepatocellular carcinoma. Cell. Death Dis. 13 , 223 (2022).

Kahl, I. et al. The cell cycle-related genes RHAMM, AURKA, TPX2, PLK1, and PLK4 are associated with the poor prognosis of breast cancer patients. J. Cell. Biochem. 123 , 581–600 (2022).

Zhao, F. et al. Identification of sixteen metabolic genes as potential biomarkers for colon adenocarcinoma. J. BUON 26 , 1252–1259 (2021).

PubMed   Google Scholar  

Yang, F. et al. Identification of key genes associated with papillary thyroid microcarcinoma characteristics by integrating transcriptome sequencing and weighted gene co-expression network analysis. Gene 811 , 146086 (2022).

Liu, Y. et al. FHL1 Inhibits the progression of colorectal cancer by regulating the Wnt/β-catenin signaling pathway. J. Cancer 12 , 5345–5354 (2021).

Eshibona, N. et al. Upregulation of FHL1, SPNS3, and MPZL2 predicts poor prognosis in pediatric acute myeloid leukemia patients with FLT3-ITD mutation. Leuk. Lymphoma. https://doi.org/10.1080/10428194.2022.2045594 (2022).

Niu, C. et al. Downregulation and growth inhibitory role of FHL1 in lung cancer. Int. J. Cancer 130 , 2549–2556 (2012).

Fang, A. et al. RAMP3 is a prognostic indicator of liver cancer and might reduce the adverse effect of TP53 mutation on survival. Future Oncol. 14 , 2615–2625 (2018).

Zhang, J., Shang, L., Jiang, W. & Wu, W. Shikonin induces apoptosis and autophagy via downregulation of pyrroline-5-carboxylate reductase1 in hepatocellular carcinoma cells. Bioengineered 13 , 7904–7918 (2022).

Oudaert, I. et al. Pyrroline-5-carboxylate reductase 1: A novel target for sensitizing multiple myeloma cells to bortezomib by inhibition of PRAS40-mediated protein synthesis. J. Exp. Clin. Cancer Res. 41 , 45 (2022).

Wu, Y. et al. A mitochondrial dysfunction and oxidative stress pathway-based prognostic signature for clear cell renal cell carcinoma. Oxid. Med. Cell Longev. 2021 , 9939331 (2021).

Albakri, M. M., Huang, S.C.-C., Tashkandi, H. N. & Sieg, S. F. Fatty acids secreted from head and neck cancer induce M2-like macrophages. J. Leukoc. Biol. https://doi.org/10.1002/JLB.1A0521-251R (2022).

Simon Davis, D. A. et al. Machine learning predicts cancer subtypes and progression from blood immune signatures. PLoS ONE 17 , e0264631 (2022).

Hecking, T. et al. Programmed cell death ligand-1 (PDL-1) correlates with tumor infiltration by immune cells and represents a promising target for immunotherapy in endometrial cancer. Anticancer Res. 42 , 1367–1376 (2022).

Zhang, L. et al. Comprehensive analysis of the MIR4435-2HG/miR-1-3p/MMP9/miR-29-3p/DUXAP8 ceRNA network axis in hepatocellular carcinoma. Discov. Oncol. 12 , 38 (2021).

Cruz, J. A. & Wishart, D. S. Applications of machine learning in cancer prediction and prognosis. Cancer Inform. 2 , 59–77 (2007).

Li, M. et al. Breath carbonyl compounds as biomarkers of lung cancer. Lung Cancer 90 , 92–97 (2015).

Zhang, J. et al. 5-Hydroxymethylome in circulating cell-free DNA as a potential biomarker for non-small-cell lung cancer. Genom. Proteom. Bioinformat. 16 , 187–199 (2018).

Zhang, Y.-H., Jin, M., Li, J. & Kong, X. Identifying circulating miRNA biomarkers for early diagnosis and monitoring of lung cancer. Biochim. Biophys. Acta Mol. Basis Dis. 1866 , 165847 (2020).

Wang, Y. et al. Screening key lncRNAs for human lung adenocarcinoma based on machine learning and weighted gene co-expression network analysis. Cancer Biomark. 25 , 313–324 (2019).

Download references

Acknowledgements

We are grateful to the GEO database for providing the platform and to the contributors for uploading their meaningful datasets. We thank Bullet Edits Limited for the linguistic editing and proofreading of the manuscript. This work was done at the First Affiliated Hospital of Guangxi Medical University.

This research was supported by Guangxi Natural Science Foundation under Grant NO. 2020GXNSFDA238003.

Author information

Authors and affiliations.

Department of Respiratory Medicine, The First Affiliated Hospital of Guangxi Medical University, Nanning, 530021, Guangxi, China

Fangwei Wang & Chaoqian Li

Department of Clinical Laboratory, The First Affiliated Hospital of Guangxi Medical University, Nanning, 530021, Guangxi, China

You can also search for this author in PubMed   Google Scholar

Contributions

W.F.W. and C.Q. L. designed the implementation of the research, drafted preliminary papers, and participated in investigations. W.F.W., Q.S.S. and C.Q.L. participated in research design and implementation, manuscript revision, manuscript submission, and fund acquisition. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Chaoqian Li .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Supplementary information., rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Wang, F., Su, Q. & Li, C. Identidication of novel biomarkers in non-small cell lung cancer using machine learning. Sci Rep 12 , 16693 (2022). https://doi.org/10.1038/s41598-022-21050-5

Download citation

Received : 06 July 2022

Accepted : 22 September 2022

Published : 06 October 2022

DOI : https://doi.org/10.1038/s41598-022-21050-5

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

This article is cited by

Machine learning pipeline to analyze clinical and proteomics data: experiences on a prostate cancer case.

• Patrizia Vizza
• Federica Aracri
• Giuseppe Tradigo

BMC Medical Informatics and Decision Making (2024)

Transcriptome profiling and metabolic pathway analysis towards reliable biomarker discovery in early-stage lung cancer

• Muthu Kumar Thirunavukkarasu
• Priyanka Ramesh
• Shanthi Veerappapillai

Journal of Applied Genetics (2024)

ABCA8 Elevation Predicts the Prognosis and Exerts the Anti-oncogenic Effects on the Malignancy of Non-small Cell Lung Cancer via TCF21-Mediated Inactivation of PI3K/AKT

• Guoqiong Zhou

Molecular Biotechnology (2023)

By submitting a comment you agree to abide by our Terms and Community Guidelines . If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing: Translational Research newsletter — top stories in biotechnology, drug discovery and pharma.

Facility for Rare Isotope Beams

At michigan state university.

Welcome to FRIB

The Facility for Rare Isotope Beams (FRIB) at Michigan State University (MSU) is a world-class research, teaching and training center, hosting what is designed to be the most powerful rare isotope accelerator. MSU operates FRIB as a user facility for the U.S. Department of Energy Office of Science (DOE-SC), supporting the mission of the DOE-SC Office of Nuclear Physics. FRIB allows MSU graduate students to engage in groundbreaking research in tandem with their coursework. Open the doors to discovery with the newest and most advanced rare isotope research facility and the world's most powerful rare isotope accelerator. Apply and inquire through FRIB’s graduate studies page at frib.msu.edu/grad .

Tentative Thesis Title: Commissioning the Separator for Capture Reactions (SECAR) for Proton Capture Reactions

Machenzie Smith

• Read more about Tentative Thesis Title: Commissioning the Separator for Capture Reactions (SECAR) for Proton Capture Reactions

Tentative Thesis Title: Precision Mass Measurement of Proton Halo Candidate 22Al and the Development of the Single Ion Penning Trap

Scott Campbell

• Read more about Tentative Thesis Title: Precision Mass Measurement of Proton Halo Candidate 22Al and the Development of the Single Ion Penning Trap

Probing Dense Matter in Neutron Stars

Juergen Schaffner-Bielich

• Read more about Probing Dense Matter in Neutron Stars

Quantum Algorithms for Simulating Nuclear Effective Field Theories

James Watson

• Read more about Quantum Algorithms for Simulating Nuclear Effective Field Theories

STREAMLINE Symposium

• Read more about STREAMLINE Symposium

In the News

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• J Clin Transl Res
• v.7(3); 2021 Jun 26

Oligometastatic non-small cell lung cancer: Current management

Alicia román-jobacho.

1 Department of Radiation Oncology, Hospital Virgen de la Victoria, Málaga, Spain

María Hernández-Miguel

2 Hospital Rey Juan Carlos, Móstoles, Madrid

María Jesús García-Anaya

Jaime gómez-millán, j. a. medina-carmona, ana otero-romero, background:.

In the past decade, major developments have improved the survival of patients with oligometastatic non-small cell lung cancer (NSCLC). About 20% - 50% of patients with NSCLC present with oligometastases at diagnosis. For this group of patients, it seems that an increase in survival would justify aggressive local therapies. The development of minimally invasive surgery and advanced radiotherapy techniques like stereotactic body radiation therapy (SBRT) makes local control possible for selected patients with metastatic NSCLC. The advantage of SBRT over surgery is that it is a non-invasive technique, with minimum side effects, and is more suitable for fragile and elderly patients, non-candidates for surgery, or patients who refuse surgery.

The purpose of this review is to summarize the latest scientific evidence on the management of oligometastatic NSCLC, focusing on the role of radiotherapy.

Relevance for Patients:

The initial treatment recommended for patients with oligometastatic NSCLC is systemic therapy. Patients should be considered for radical treatment to both the primary tumor and oligometastases. Aggressive local therapy comprises surgery and/or definitive radiotherapy such as SRS or SBRT, and may be preceded or followed by systemic treatment. Recent clinical evidence from Phase II trials reports benefits in terms of PFS in patients with good performance status and long disease-free periods, with good response to systemic therapy, especially in EGFR wild-type tumors. Phase I and II trials have shown that radiotherapy combined with immunotherapy can improve tumor response rate and possibly overall survival. The recommendation is also to include OM patients in ongoing clinical trials.

1. Introduction

Stage IV non-small cell lung cancer (NSCLC) has traditionally been considered an incurable disease, with low life expectancy, even when treated with standard chemotherapy based on platinum. However, in the past decade, major advances have improved the survival of these patients, especially those with oligometastatic NSCLC (OM-NSCLC)

In 1995, Helmann and Weichselbaum were the first to describe the oligometastatic state as an intermediate state between localized disease and widespread disease, with a more indolent biology and lower dissemination capacity [ 1 ]. The 8 th edition of American Joint Committee on Cancer reclassified Stage IV NSCLC based on the number of metastases, with M1b an extrathoracic metastasis (IVA) and M1c multiple extrathoracic metastases in one or more organs (IVB) [ 2 ].

This new classification divides groups of patients according to different therapeutic approaches and prognoses. In NSCLC, the oligometastatic state is relatively common, with 20–50% of patients having oligometastatic disease at diagnosis.

It seems that for this group of patients, an increase in survival would justify aggressive local therapies [ 3 ].

Initially, the only radical treatment for oligometastases was surgery. The development of minimally invasive surgery and advanced radiotherapy techniques like stereotactic body radiation therapy (SBRT) makes local control (LC) possible for selected patients with metastatic NSCLC. SBRT allows the treatment of lesions in any part of the body, with a small number of fractions and a minimum toxicity profile. It also has an advantage over surgery in that it is a non-invasive technique, with minimum side effects, and is more suitable for fragile and elderly patients, non-candidates for surgery, or patients who refuse surgery.

In an individual patient meta-analysis of 757 patients with NSCLC, Ashworth et al . found that most oligometastases were in the brain (35.5%) or lung (33.6%), followed by the adrenal gland (13.0%), bone (8.5%), other (7.8%), liver (2.4%), and lymph node (2.4%) [ 4 ]. The meta-analysis revealed that the median overall survival (OS) in patients treated with ablation of all disease sites including the primary tumor was 26 months, and survival at 1, 2, 5, and 8 years was 70.2%, 51.1%, 29.4%, and 23.4%, respectively [ 5 ].

2. Definitions of Oligometastases

There is no well-established number of metastases to define a patient as oligometastatic. A European multidisciplinary consensus statement on the definition and staging of OM-NSCLC considered that the maximum number of metastases/organs involved depends on the possibility of offering a radical-intent treatment strategy. Based on the systematic review, a maximum of five metastases and three organs was agreed on. The presence of diffuse serosal metastases or bone marrow involvement excludes cases from this definition [ 6 ].

There are additional suggested definitions that cover different scenarios in the presentation of oligometastases.

• “ De novo oligometastases” or “synchronous oligometastasis ” refers to the presence of a limited number of metastases at diagnosis. Patients with widely disseminated disease at diagnosis who present with oligometastases after systemic treatment are said to have “induced oligometastases” [ 1 ]
• The term oligorecurrence refers to progression with new metachronous oligometastases after definitive treatment of the primary locoregional thoracic disease [ 7 ]
• Oligopersistent disease is a concept that refers to oligometastatic patients who remain stable after systemic therapy or who, starting from a more widespread disease, achieve an oligometastatic state [ 8 ]
• Oligoprogressive disease describes a situation in which patients with disseminated disease at diagnosis respond to systemic treatment, remaining stable while one or a limited number of metastases progress during systemic therapy [ 9 ].

The frequency of oligoprogression during treatment with tyrosine kinase inhibitors (TKIs) varies according to the definition used, however, estimates vary from 15% to 47% [ 10 , 11 ]. It is believed that oligoprogression arises as a result of tumor heterogeneity and the development of isolated resistance in one or more metastatic sites [ 12 ].

The prognosis among these clinical situations is different, and each represents a heterogeneous group contingent on the number of metastatic lesions, tumor genotype (EGFR mutated, ALK rearrangements, and so forth), and type of systemic treatment [ 13 ].

3. Local Ablative and/or Systemic Therapy to Synchronic Oligometastatic State

Two randomized Phase II trials have shown an increased PFS by adding radical local treatment to systemic therapy in patients with oligometastatic NSCLC who achieve good response [ 14 , 15 ].

Nevertheless, many questions remain regarding the treatment of these patients, which patients receive the greatest benefit from ablative treatment.

If not previously performed, Stage IV NSCLC patients should undergo evaluation with brain MRI or CT scan and whole-body PET scan to ensure oligometastatic status. Although most studies were conducted before the PET scan era, recommendations are to perform such an evaluation because approximately 15% of patients with NSCLC initially classified as Stage I-III by CT scan will change classification to Stage IV with a PET scan [ 16 ].

National Comprehensive Cancer Network guidelines include the recommendation of a confirmatory biopsy of the metastatic lesion whenever possible [ 4 , 17 ].

In a recent multi-institutional Phase II study, Gomez et al . observed an improvement in OS after aggressive local treatment of NSCLC patients with three or fewer metastatic lesions that had not progressed to the first line of chemotherapy. The trial closed early due to the statistically significant median progression-free survival advantage for patients in the local consolidation arm of 14.2 months versus 4.4 months. Later, long-term results showed a significantly longer median OS for these patients (41.4 months vs. 17 months) [ 14 ].

A subsequent randomized Phase II trial showed similar favorable results in 29 patients with five or fewer metastases and partial response or stability after systemic treatment. All lesions, including primary lesions, were treated with SBRT or hypofractionated radiation therapy followed by maintenance chemotherapy versus maintenance alone [ 15 ]. This trial also closed early due to better disease-free survival (DFS) in the consolidated local therapy arm, with a median DFS of 9.7 months compared to 3.5 months. In addition, no patient with local treatment developed progression in an irradiated site, as opposed to 70% in the exclusive maintenance group.

Surgery is an established treatment for cerebral, pulmonary, and adrenal oligometastases of lung cancer.

Occasionally, patients with isolated metastases in other locations (e.g. bone and liver) have received treatment with metastasectomy; however, the number reported is much smaller as this treatment may not be suitable for patients with several affected locations. Altogether, systemic therapies alone may not be optimal in disease settings such as oligometastatic lung cancer, where long-term control can be expected. Kissel et al . assessed the outcomes of 91 lung cancer patients with extracranial metastases in oligometastatic, oligorecurrent, oligopersistent, and oligoprogressive settings (the “oligometastatic spectrum”) under strategies using SBRT ± systemic treatments. With a median follow-up of 15.3 months, crude LC at irradiated metastases was 91%, whereas median distant progression-free survival (dPFS) oligoprogressive patients had the worst dPFS, and OS was 6.3 and 28.4 months, respectively (2-year survival 54%). Initial nodal stage and the oligometastatic spectrum were prognostic factors for dPFS; the multivariate analysis showed age, initial primary stage, and oligometastatic spectrum as prognostic factors for survival. For oligorecurrent patients, a longer free interval between primary and metastatic spread was associated with better survival, with a threshold of 2 years ( P =0.002) [ 18 ].

Options for local therapy include surgery, SBRT, and radiofrequency. Due to the lack of randomized trials that compare these modalities, the choice of local therapy in oligometastases depends on several factors and a multidisciplinary team should take this decision based on:

• Patient-related factors including performance status, age, comorbidities, and patient decision
• Metastases-related factors such as size, number of lesions, resectability, and remaining volume of site organ
• Treatment-related factors: Technique availability, professional training, and costs.

There are several ongoing Phase II/III trials to assess the role of SBRT treatment in oligometastatic NSCLC combined with chemotherapy and/or immunotherapy. Results will be available in the near future, Table 1 [ 19 ].

SBRT: Stereotactic body radiotherapy; LCT: Local consolidative therapy; NSCLC: Nonsmall cell lung cancer

No research reports direct comparisons between surgery and radiotherapy as a radical treatment for oligometastatic disease, but several studies have included patients treated with either surgery or radiotherapy. Evidence for the other ablative modality, radiofrequency ablation, is still limited. The advantages of radiofrequency ablation are the freedom to perform the procedure regardless of any previous therapy and its repeatability. However, rare but serious complications may occur, including bronchopleural fistula, pulmonary artery pseudoaneurysm, systemic air embolism, injury of the brachial nerve and phrenic nerve, pneumonia, and needle tract seeding of the cancer [ 13 ].

The ongoing Phase II STOP trial includes patients with NSCLC in oligoprogression and randomizes them to receive their standard systemic treatment or SBRT to all oligoprogression sites (maximum five). The primary endpoint is progression-free survival, and the expected completion date of this study is in 2021 [ 20 ].

4. Treatment of the Primary Tumor

Treatment of the primary tumor is a topic of debate. Justification for a radical approach to the primary tumor could be that distant spread of the tumor comes from both the primary tumor and metastases, so radical treatment to both could improve OS in OM NSCLC patients [ 21 ].

It is known that patients with a solitary site of metastasis, frequently the brain or adrenal gland, who undergo surgical resection of both the primary and the metastasis can occasionally experience long-term survival or even cure [ 22 ].

Petrelli et al . conducted a systematic review and meta-analysis to study the role of local radiation therapy to the primary tumor in synchronous OM NSCLC [ 23 ]. Their review included 21 studies: Four Phase II studies, two randomized Phase II studies, and 15 retrospective series. Radiotherapy was delivered with palliative doses in 1.4% of patients, the rest of cases received a treatment of 3D external radiotherapy or SBRT, either alone (21%), in combination with sequential or concomitant chemotherapy (79%), or adjuvant to surgery (0.4%). The median pooled OS was 20.4 months (95% CI 16.6–24.7) and median PFS was 12 months (95% CI 9.8–14.5). Thoracic radiation therapy improved OS (HR=0.44, 95% CI 0.32–0.6; P <0.001) and PFS (HR=0.42, 95% CI 0.33–0.55; P <0,001).

Li et al . conducted a meta-analysis to identify prognostic factors of OM NSCLC and found that the factors associated with better outcomes using a radical approach to the primary tumor were sex (female), (y)pT stage, negative nodal status, and adenocarcinoma histology [ 24 ].

The Phase II trial by Gomez et al . mentioned above compared outcomes in 49 OM NSCLC patients (≤3 metastases) with no progression after a minimum 4 cycles of platinum-based chemotherapy or 3 months of anti-EGFR/anti_ALK therapy. Patients were randomized to receive local radiotherapy followed by maintenance systemic therapy or observation versus only maintenance or observation. The trial found a significant benefit in median progression-free survival (PFS) in the consolidative local treatment arm (14.2 months vs. 4.4 months). Long-term results published in 2019 reported significantly longer median OS in consolidative local treatment (41.2 months vs. 17.0 months). Regarding local treatment, 67% of patients received radiation therapy, either concomitant with chemotherapy, exclusively hypofractionated radiotherapy or SBRT, while 33% of patients had surgery to the primary tumor. Radiotherapy regimens varied considerably, from palliative doses to ablative doses, but even lower doses of radiation provided a benefit in delaying progression [ 14 ].

Iyengar et al . explored the role of radiation therapy to both the primary tumor and the metastases in 29 patients with OM NSCLC, excluding EGFR- or ALK-mutated tumors. Patients with stable disease or partial response after 4–6 cycles of chemotherapy were randomized to receive SBRT to the metastases and radiation therapy (SBRT or hypofractionated RT) followed by maintenance chemotherapy versus maintenance chemotherapy alone. The trial closed early due to significantly better PFS in the consolidative local treatment arm (9.7 months vs. 3.5 months) [ 15 ].

5. Combination of Systemic Therapy and Radiotherapy

The current level of evidence does not support the routine use of local ablative treatment as the initial treatment in oligometastatic disease, for which systemic therapy remains the standard of care. Local treatment approaches could be considered for patients not suitable for systemic therapy or for those who refuse or want to delay it [ 8 ].

The most suitable treatment sequence remains a matter of controversy because data come from retrospective studies or small prospective studies.

Various factors require consideration, such as the time of diagnosis (synchronous vs. metachronous), primary tumor local extension, the number, size and location of the metastases, performance status, the absence or presence of symptoms, patient comorbidities, and the absence or presence of target mutations.

One accepted strategy in patients with synchronous metastases is to begin with systemic treatment and later evaluate disease response before radical treatment to the primary tumor as well as metastases.

After morphological diagnosis, the next consideration is therapy-predictive biomarker testing. It is well known that EGFR-mutated and ALK-translocated NSCLC treated with oncogene-driven targeted therapy has significantly better outcomes compared with standard platinum-based systemic therapy [ 25 , 26 ]. Acquired resistance to first-line TKI can occur on average after 9–12 months of treatment. Patterns of progression include generalized progression, brain progression, or oligoprogression. Oligoprogression is estimated to occur during TKI treatment in 15–47% of patients due to tumor heterogeneity and an isolated resistant subclone at 1–4 metastatic sites [ 12 ]. Data from retrospective studies suggest that aggressive local treatment can eradicate TKI-resistant oligometastases offering a benefit in LC. In this setting, the preference is for SBRT rather than surgery or normofractionated radiotherapy due to its shorter duration and minimal side effects. In this manner, SBRT allows continued use of TKI to control the majority of the disease [ 27 ].

In 2018, De Ruysscher et al . [ 28 ] conducted a single-arm prospective Phase II trial with 39 NSCLC Stage IV patients who had fewer than five metastases at primary diagnosis and management with radical local treatment (radiotherapy or surgery). No previous response to systemic treatment was required. Thirty-seven patients (95%) received chemotherapy. The median OS was 13.5 months and median PFS was 12.1 months. Three patients (7%) had a local relapse.

For patients with metastatic disease limited to the central nervous system, initiation of treatment with stereotactic radiosurgery (SRS) or surgical resection is an option; however, for patients with ALK or EGFR mutation who present asymptomatic lesions, it is reasonable to start with targeted therapy before radiotherapy treatment [ 29 , 30 ].

In the metachronous setting, the Phase II SABR-COMET demonstrated a survival benefit of SBRT as a consolidation treatment in the context of oligometastatic disease. The trial recruited patients with metachronous oligometastasis disease of any histology who had previously received treatment to the primary tumor. Patients treated with SBRT to metastases sites had better OS compared to systemic treatment alone [ 31 ].

The randomized prospective Phase II/III trial, NRG-LU002, examines maintenance systemic therapy alone compared with SBRT to all metastatic sites (three or more), plus radiation to the primary site followed by systemic therapy to determine the benefit of treatment of all metastases in NSCLC. Primary outcomes include PFS and OS. The UK conventional care versus radioablation (stereotactic body radiotherapy) for extracranial oligometastases trial is a multicenter, randomized, Phase II trial investigating whether the addition of SBRT to standard therapy in patients with oligometastatic NSCLC, breast, renal cell, or prostate cancer (three or more extracranial metastases) will improve PFS, OS, LC, and freedom from widespread metastatic disease. It will also collect toxicity and quality-of-life data. These data will represent some of the first available robust randomized evidence regarding the use of metastasis-directed local therapy in the setting of oligometastatic NSCLC [ 32 , 33 ].

6. Immunotherapy and Ablative Radiotherapy

Patients with oligometastatic lung disease are commonly treated with immunotherapy, and findings show that ionizing radiation can produce greater antigen presentation to better immune system recognition [ 34 ].

Unfortunately, not all patients respond to this treatment, possibly because tumor antigens are not recognized.

Ablative doses of SBRT can induce necrosis and senescence, types of cell death in which there is a greater release of tumor antigens and higher infiltration of T lymphocytes in the irradiated tumor with the release of cytokines related to cell damage [ 35 - 37 ].

SBRT may synergize with immunotherapy; the combination of both treatments shows the production of more tumor regression in several solid tumor types.

Ionizing radiation can also increase the effect of immune checkpoint inhibitors such as pembrolizumab or durvalumab. Thus, some authors have postulated that SBRT before immunotherapy can lead to the improved efficacy of this immunological treatment [ 38 ].

Phase I and II trials have tested the safety of the SBRT and immunotherapy combination. In 2014, Tang published results of the first Phase I/Phase II trial that combined SBRT with immunotherapy in patients with advanced NSCLC. The study observed stable disease or partial response in 67% of patients, and patients who received sequential radiation to lung metastases had better OS than those who received sequential radiation to liver metastases [ 39 , 40 ].

In 2015, the KEYNOTE-001 trial demonstrated the efficacy and safety of durvalumab in advanced lung cancer. A secondary analysis found that patients who received radiotherapy prior to pembrolizumab treatment doubled OS (10.7 vs. 5.3 months) [ 41 , 42 ].

The PEMBRO-RT Phase II trial published its results in July 2019. The trial included patients with recurrent metastases after at least one regimen of chemotherapy. Patients were randomized to either pembrolizumab alone, a selective humanized PD-1 monoclonal antibody, or pembrolizumab after SBRT on a single tumor site. Patients had at least two separate metastatic lesions, only one of which was irradiated, and response was measured in both locations to test the immunological effect in the non-irradiated area. Although the overall response rate at 12 weeks was 18% in the control arm versus 36% in the experimental arm, these results did not meet the study criteria for meaningful clinical benefit. A subgroup analysis revealed that the patients who benefited the most, in terms of progression-free survival and OS, were those with tumors that did not express PD-L1, suggesting that combined treatment improved antigen presentation and response [ 43 ].

A subsequently published pooled study that analyzed the results of PEMBRO-RT together with the MDACC study [ 44 ], which had similar inclusion criteria and treatment schemes, showed the best out-of-field response rate with radiotherapy plus pembrolizumab versus pembrolizumab alone (41.7% vs. 19.7%, P : 0.0039). However, longer median progression-free survival (9 months vs. 4.4 months, P : 0.04) and significantly better median OS (19.2 months vs. 8.7 months, P : 0.0004) occurred among the radiotherapy group [ 45 ].

Hence, it would appear that adding radiotherapy to immunotherapy can increase responses and outcomes in patients with metastatic lung cancer, although validation in a randomized Phase 3 trial is necessary.

7. Treatment of Oligometastases by Site

7. 1. brain oligometastasis.

Approximately 25–30% of patients who have lung cancer present with brain metastases at diagnosis. In certain situations, such as a limited number of metastases, favorable histology, and good performance status, the radical management of brain metastases is feasible, which translates into better outcomes [ 46 , 47 ].

Brain-only oligometastases represent a subgroup with better prognosis if radical treatment to primary and cerebral lesions is possible, even in node-positive disease, and prognosis is similar to Stage III with definitive chemoradiotherapy, with a median OS <24 months and 3-year OS of 40% [ 48 ].

The most important treatment approaches in this situation are surgery, whole-brain radiotherapy (WBRT), and SRS. Clinicians do not use WBRT as often in patients with limited brain metastases due to concerns about long-term toxicity, which includes neurocognitive effects.

7. 2. Adrenal gland oligometastasis

Adrenalectomy was the first treatment modality used in the management of adrenal metastases, although SBRT is beginning to offer a safe and effective alternative in non-operable cases.

Current evidence comes from retrospective studies, and the number of fractions and doses prescribed varies among the series (10–45 Gy in 3–10 fractions) [ 13 ].

Chen et al . issued an analysis of 39 studies published between 2009 and 2019, reporting outcomes of 1006 patients. The pooled 1- and 2-year rates of LC were 82% (74–88%) and 63% (50–74%), respectively, and the pooled 1- and 2-year OS rates were 66% (57–74%) and 42% (31-53%), respectively. There was a strong positive association between SBRT dose and 1- and 2-year LC ( P <0.0001 and P =0.0002), and an association with 2-year OS ( P =0.03). The overall rate of Grade 3 or higher toxicity was 1.8%. They concluded that SBRT for adrenal metastases was a safe treatment associated with excellent 1-year LC, the effective palliation of pain and a reduction of tumor volume. The ideal dose is still under investigation [ 49 ].

7. 3. Liver oligometastases

At present, surgery is the local treatment of choice for liver metastases and it achieves a LC of 85–90%; however, only 10-20% of patients with liver metastases are candidates for surgery [ 13 ]. There are few reported cases of liver metastasis resection from lung cancer, but long-term survival is high (21–60 months), which can be explained by patient selection [ 50 , 51 ].

Despite surgery being contraindicated for most liver metastases from lung cancer, radiotherapy can be a local ablative treatment option when indicated.

Many prospective and retrospective studies [ 52 ] have analyzed SBRT in liver oligometastases. All of the studies included patients with liver metastases from different primary tumors. LC at 2 years was around 90% and had a low toxicity profile, doses ranged from 36 to 60 Gy in 3 to 6 sessions. Rusthoven’s Phase II study showed an OS of 12 months for the unfavorable group [ 53 - 55 ].

With the scientific evidence currently available, high LC rates with SBRT would seem to be similar to those obtained using other local treatments, however, the results of the prospective studies currently in progress may alter these findings.

Thus, patients with 1–3 hepatic lesions can benefit from aggressive local treatment with SBRT to improve LC and possibly improve survival.

7. 4. Lung oligometastases

The metastasectomy study based on 5206 cases of different histologies published its results in 1997 [ 56 ]. With a mean follow-up of 46 months, actuarial survival at 5, 10, and 15 years was 36%, 26%, and 22%, respectively.

However, only a few patients were candidates for surgery. After lobectomy, LC rates ranged from 85% to 95%, and after wedge resections, LC rates ranged from 50% to 70% [ 57 , 58 ].

Phase I and II trials of SBRT for primary tumors and lung oligometastases have demonstrated feasibility, safety, and efficacy with good to excellent LC in most studies. LC rates reported using SBRT are in the range of 70–90% at 2 years, which are similar to those obtained with metastasectomy [ 59 ].

Rieber et al . reviewed 700 patients with medically inoperable lung metastases secondary to different malignancies treated with SBRT in 20 German centers between 1997 and 2014. They evaluated primary and metastatic tumor characteristics, treatment characteristics, and follow-up data including survival, LC, distant metastases, and toxicity. Lung metastases were treated with median PTV-encompassing single doses of 12.5 Gy (range 3.0–33.0 Gy) in a median number of three fractions (range 1–13). Two-year LC and OS were 81.2% and 54.4%, respectively. Independent prognostic factors for LC were performance status and biological effective dose at both isocenter and periphery. Survival was significantly better for patients with a good performance status, small and single pulmonary metastases, a long time interval between primary tumor diagnosis and SBRT treatment, and a favorable primary tumor histology [ 60 ].

Ashworth et al . carried out a meta-analysis of 757 patients with NSCLC with 1–5 synchronous or metachronous metastases treated with surgical metastasectomy, stereotactic radiotherapy/radiosurgery, or radical external beam radiotherapy, and curative treatment of the primary lung cancer, from hospitals worldwide. They observed significant OS differences in oligometastatic patients depending on the time of the oligometastatic presentation; thus, survival was better in metachronous disease than in synchronous disease ( P <0.001) [ 5 ].

At present, there are no randomized trials that compare SBRT and surgery in the treatment of lung metastases, however, both treatments seem to have similar LC.

7. 5. Other locations

SBRT for other locations of oligometastases, such as bone, spinal cord, and lymph nodes, has been gaining widespread acceptance given the low toxicity profile and excellent LC rates reported in retrospective studies, which are as high as 85-100%. Doses administered with SBRT vary among the different series, making comparisons difficult. Prospective studies are necessary to clarify the most appropriate dose and timing of SBRT in these locations [ 61 , 62 ].

8. Conclusion

In general, the recommendation is for systemic therapy as the initial treatment for patients with oligometastatic NSCLC. Aggressive local therapy comprises surgery and/or definitive radiotherapy such as SRS and SBRT, and may be preceded or followed by systemic treatment. Patients should be considered for radical treatment to both the primary tumor and oligometastases. The selection of patients who might clearly benefit from a radical approach is challenging, and a multidisciplinary team should discuss such a decision. Recent clinical evidence from Phase II trials reports benefits in terms of PFS in patients, with good performance status and long disease-free periods, and a good response to systemic therapy, especially in EGFR wild-type tumors.

Phase I and II trials have shown that radiotherapy combined with immunotherapy can improve tumor response rate and possibly OS. The recommendation is also to include OM patients into ongoing clinical trials.

Conflicts of interest

All the authors declare that there are no conflicts of interest.

< < return to the NSCL Library main page

• Bibliography
• More Referencing guides Blog Automated transliteration Relevant bibliographies by topics
• Automated transliteration
• Relevant bibliographies by topics
• Referencing guides

Dissertations / Theses on the topic 'NSCL'

Create a spot-on reference in apa, mla, chicago, harvard, and other styles.

Consult the top 50 dissertations / theses for your research on the topic 'NSCL.'

Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.

You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.

Browse dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.

Krüger, Marcus. "Molekulare und funktionale Charakterisierung der neuronalen bHLH Transkritptionsfaktoren NSCL-1 und NSCL-2." [S.l.] : [s.n.], 2002. http://deposit.ddb.de/cgi-bin/dokserv?idn=969337183.

Pegoraro, Ilan Emanuel Moreira. "NSCL in the ETSI M2M platform." Master's thesis, Universidade de Aveiro, 2014. http://hdl.handle.net/10773/14704.

Daskalos, Alexandros. "Deregulation of DNA methylation aand retrotransposon reactivation in NSCL." Thesis, University of Liverpool, 2011. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.539561.

Richard, Andrea L. "Spectroscopy of the A = 33 Isobars in the Island of Inversion." Ohio University / OhioLINK, 2018. http://rave.ohiolink.edu/etdc/view?acc_num=ohiou1519055652016583.

Howard, Meredith E. "The Joy of CEX: Sharpening the ( t , 3 He) probe at 345 MeV for the charge-exchange knife drawer." Columbus, Ohio : Ohio State University, 2008. http://rave.ohiolink.edu/etdc/view?acc%5Fnum=osu1222168015.

Divaratne, Dilupama A. "One and Two Neutron Removal Cross Sections of 24 O via Projectile Fragmentation." Ohio University / OhioLINK, 2014. http://rave.ohiolink.edu/etdc/view?acc_num=ohiou1395839092.

Stone, Michael H. "North Carolina State NSCA Meeting." Digital Commons @ East Tennessee State University, 2005. https://dc.etsu.edu/etsu-works/4488.

Morais, Christianni Cardoso. "Para aumento da instrução da mocidade da nossa pátria: estratégias de difusão do letramento na Vila de São João Del Rei (1824-1831)." Universidade Federal de Minas Gerais, 2002. http://hdl.handle.net/1843/NSCS-5NGJJF.

Klinke, Karina. "Escolarização da leitura no ensino primário de Minas Gerais (1906-1930)." Universidade Federal de Minas Gerais, 2003. http://hdl.handle.net/1843/NSCS-5T6F6P.

Santos, Leonardo Alberto de Azevedo. "A presença do pensamento de Antonio Gramsci na produção acadêmica sobre educação no Brasil dos anos oitenta." Universidade Federal de Minas Gerais, 2002. http://hdl.handle.net/1843/NSCS-5NGMUP.

Silva, Marcilene da. "Indios civilizados e escolarizados: a produção de uma outra condição de etnicidade." Universidade Federal de Minas Gerais, 2003. http://hdl.handle.net/1843/NSCS-5T6FW2.

Resende, Tania de Freitas. "Escola tá bravo... bravo, escola!: acesso à informação fora da escola e construção dos conhecimentos escolares em sala de aula." Universidade Federal de Minas Gerais, 2003. http://hdl.handle.net/1843/NSCS-5TENGW.

Santo, Shirlei Rezende Sales do Espirito. "Oposição, diversão e violência na escola: os significados produzidos para práticas culturais de transgressão." Universidade Federal de Minas Gerais, 2002. http://hdl.handle.net/1843/NSCS-5NGMSZ.

Chater, Emily. "Novel therapeutic targets in NSCLC resistance to Erlotinib." Thesis, Imperial College London, 2017. http://hdl.handle.net/10044/1/50699.

Holgersson, Georg. "Prognostic Factors in Non-Small Cell Lung Cancer (NSCLC)." Doctoral thesis, Uppsala universitet, Experimentell och klinisk onkologi, 2017. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-327925.

Anajafimarzijarani, Hamidreza. "Improved Seismic Design of Non-structural Components (NSCs) and Development of Innovative Control Approaches to Enhance the Seismic Performance of Buildings and NSCs." Thesis, University of New Hampshire, 2018. http://pqdtopen.proquest.com/#viewpdf?dispub=10935426.

Post-earthquake reconnaissance following past earthquakes in the US and other seismic-prone countries illustrates that the majority of building losses (injury, dollar loss and downtime) resulted from damage to nonstructural components (NSCs) and building contents. NSCs damages can severely compromise a building functionality, even if the building does not suffer significant structural damages. NSCs can be classified either as primarily displacement/deformation-sensitive or acceleration-sensitive. This study focuses on acceleration-sensitive components.

Previous studies on NSCs are mostly based on the responses of simplified models of primary systems and components. These models, while providing valuable insight into understanding the influential parameters on NSCs seismic demands and behavioral patterns, may not adequately represent the characteristics present in the response of actual buildings. In the first part of this dissertation, acceleration responses of a wide variety of instrumented buildings and code-based designed building models are evaluated to: ( i ) identify the most important limitations of using simplified numerical models, ( ii ) quantify the most influential parameters that control NSC responses, ( iii ) evaluate the design equivalent static equations of ASCE 7-16 for acceleration-sensitive NSCs, ( iv ) assess alternative design equivalent static equations proposed as part of a recent project sponsored by the Applied Technology Council (Project ATC-120), and ( v ) develop modifications and improvements to the proposed ATC-120 equations.

In the second part of this dissertation, modern seismic protection techniques are studied that can decrease seismic input demands to a building, as opposed to modifying the seismic resistance of a building, which is the approach taken in current US design seismic provisions. The conventional base-isolation and tuned-mass-damper concepts are utilized to develop an innovative seismic control system (i.e., partial mass isolation, PMI) that can reliably enhance the seismic performance of the structural elements and NSCs so that the building can be occupied and remain functional immediately after a design earthquake. The practicality, limitations, effectiveness, and robustness of the PMI system for protecting the structural and nonstructural components of building structures are discussed and evaluated.

Exter, Kristin den. "Integrating environmental science and management the role of system dynamics modelling /." Connect to this title online, 2004. http://thesis.scu.edu.au/adt-NSCU/public/adt-NSCU20041214.143956/.

Coleman, Anita Sundaram, and Youfen Su. "The NSDL as a testbed for digital library learning research." Project Kaleidoscope, 2004. http://hdl.handle.net/10150/105461.

Chatterjee, Saradiya. "Role of TLR7 in non-small cell lung carcinona (NSCLC)." Paris 6, 2013. http://www.theses.fr/2013PA066063.

Recondo, Gonzalo. "Resistance Mechanisms to ALK Tyrosine Kinase Inhibitors (TKIs) in NSCLC." Thesis, Université Paris-Saclay (ComUE), 2019. http://www.theses.fr/2019SACLS248/document.

Stamatkin, Christopher W. "PHOSPHATIDYLINOSITOL 3-KINASE (PI3K) AS A THERAPEUTIC TARGET IN NSCLC." UKnowledge, 2014. http://uknowledge.uky.edu/pharmacy_etds/58.

Baghai, Tabassom. "ATF3 as a Key Regulator of Cisplatin Cytotoxicity: Combining ATF3 Inducing Agents Enhances Cisplatin Activity in NSCLC." Thesis, Université d'Ottawa / University of Ottawa, 2018. http://hdl.handle.net/10393/37963.

Guo, Duoli. "SELECTED STUDIES IN PHARMACEUTICS." Diss., The University of Arizona, 2010. http://hdl.handle.net/10150/195941.

Pereira, Maria Selma da Camara Lima. "Competência lingüística em pré-escolares de NSC distintos." reponame:Repositório Institucional da UFSC, 2013. https://repositorio.ufsc.br/handle/123456789/106030.

Souza, Cristiano de Pádua. "Perfil de expressão de microRNAs e seus alvos moleculares em carcinoma pulmonar." Botucatu, 2016. http://hdl.handle.net/11449/140150.

Campbell, Thomas. "The role of voltage-gated sodium channels in non-small cell lung cancer." Thesis, University of Manchester, 2013. https://www.research.manchester.ac.uk/portal/en/theses/the-role-of-voltagegated-sodium-channels-in-nonsmall-cell-lung-cancer(a65f4c5e-b217-483b-91d3-bb669965eb03).html.

Sarin, Navin [Verfasser]. "Cisplatin resistance is associated with altered signalling in NSCLC cells / Navin Sarin." Bonn : Universitäts- und Landesbibliothek Bonn, 2018. http://d-nb.info/1153467119/34.

Schaal, Courtney. "Regulation of nAChRs and Stemness by Nicotine and E-cigarettes in NSCLC." Scholar Commons, 2016. http://scholarcommons.usf.edu/etd/6582.

Sandi, Chiranjeevi. "Investigating the pathogenesis and therapy of Friedreich ataxia." Thesis, Brunel University, 2010. http://bura.brunel.ac.uk/handle/2438/6334.

Petroff, Alev [Verfasser], and Christian [Akademischer Betreuer] Rübe. "Langzeitergebnisse radikal behandelter synchron vs. metachron oligometastasierter NSCLC / Alev Petroff ; Betreuer: Christian Rübe." Saarbrücken : Saarländische Universitäts- und Landesbibliothek, 2019. http://d-nb.info/1175950270/34.

FORNI, MARCEL DE. "Essai clinique phase i du dup 785 (nsc 368390) conduit a l'institut gustave roussy sur le schema d'une courte administration intra-veineuse 5 jours consecutifs (protocole 002c1) : analyse intermediaire ; octobre 1985 - decembre 1987." Toulouse 3, 1988. http://www.theses.fr/1988TOU31143.

Koch, Franziska [Verfasser]. "Humanes ex-vivo Lungentumormodell : Nutzung für Temperaturmessungen während der Thermoablation von NSCLC. / Franziska Koch." Greifswald : Universitätsbibliothek Greifswald, 2011. http://d-nb.info/1018242198/34.

Balko, Justin M. "THE PHARMACOGENOMICS OF EGFR-DEPENDENT NSCLC: PREDICTING AND ENHANCING RESPONSE TO TARGETED EGFR THERAPY." Lexington, Ky. : [University of Kentucky Libraries], 2009. http://hdl.handle.net/10225/1062.

Parikh, Ravi B. "DEFINITIVE PRIMARY THERAPY IN PATIENTS PRESENTING WITH OLIGOMETASTATIC NON-SMALL CELL LUNG CANCER (NSCLC)." Thesis, Harvard University, 2014. http://etds.lib.harvard.edu/hms/admin/view/47.

Ramos, Alexis. "Cancer Genome Characterization with SNP Array and Whole-Exome Sequencing Analysis." Thesis, Harvard University, 2011. http://dissertations.umi.com/gsas.harvard:10036.

Schneeberger, Valentina. "Novel Roles of the Protein Tyrosine Phosphatase SHP2 in Non-small Cell Lung Cancer." Scholar Commons, 2014. https://scholarcommons.usf.edu/etd/5306.

Usó, Marco Marta. "ANALYSIS OF IMMUNOREGULATORY BIOMARKERS IN NON-SMALL CELL LUNG CANCER." Doctoral thesis, Universitat Politècnica de València, 2015. http://hdl.handle.net/10251/51283.

Prokkola, J. (Jarmo). "Enhancing the performance of ad hoc networking by lower layer design." Doctoral thesis, University of Oulu, 2008. http://urn.fi/urn:isbn:9789514289323.

Heinrich, Aaron David. "Delayed neutrons from the neutron irradiation of ²³⁵U." Texas A&M University, 2008. http://hdl.handle.net/1969.1/85943.

Burger, A. M., Paul M. Loadman, D. E. Thurston, R. Schultz, H. H. Fiebig, and Michael C. Bibby. "Preclinical pharmacology of the pyrrolobenzodiazepine (PBD) monomer DRH-417 (NSC 709119)." Italian Society of Chemotherapy and the Italian Federation of Human and Animal Mycopathology, 2007. http://hdl.handle.net/10454/4571.

Reuter, Cécile [Verfasser]. "Mutationsabhängige Aktivität von niedermolekularen reversiblen und irreversiblen Inhibitoren der EGFR Signalkaskade in NSCLC / Cécile Reuter." Köln : Deutsche Zentralbibliothek für Medizin, 2012. http://d-nb.info/1026215080/34.

Reinert, Christian Philipp [Verfasser]. "Einfluss von Lipoteichonsäuren von Staphylococcus aureus auf die Proliferation von NSCLC-Zelllinien / Christian Philipp Reinert." Gießen : Universitätsbibliothek, 2017. http://d-nb.info/1130119807/34.

Haggerty, Leonard. "A Profile of Strength and Conditioning Coaches at National Collegiate Athletic Association Division II and III Member Institutions." Digital Commons @ East Tennessee State University, 2005. https://dc.etsu.edu/etd/1094.

Paoletti, Andrew Michael. "DELAYED OR MISSING PREVENTATIVE CARE AND PROBLEMS GETTING SPECIALIST CARE IN CHILDREN WITH MENTAL HEALTH DISORDERS." Master's thesis, Temple University Libraries, 2017. http://cdm16002.contentdm.oclc.org/cdm/ref/collection/p245801coll10/id/440758.

Hoang, Thi Minh Phuong Thi minh Phuong. "Optimisation des temps de calculs dans le domaine de la simulation par éléments discrets pour des applications ferroviaires." Thesis, Montpellier 2, 2011. http://www.theses.fr/2011MON20212/document.

川合, 隆光, TAKAMITSU KAWAI, 周一 市川, SHUICHI ICHIKAWA, 俊夫 島田, and TOSHIO SHIMADA. "並列数値シミュレーション用高水準言語NSL." 情報処理学会, 1997. http://hdl.handle.net/2237/10312.

Tam, Yee-san Issan, and 譚薏珊. "Epidermal growth factor receptor (EGFR) mutations and phosphorylation pattern in non-small cell lung cancer (NSCLC)." Thesis, The University of Hong Kong (Pokfulam, Hong Kong), 2008. http://hub.hku.hk/bib/B40687673.

Bertram, Veit [Verfasser], and Rainer M. [Akademischer Betreuer] Bohle. "Quantitative Expressionsanalyse von TTF-1 in NSCLC und pulmonalen Metastasen / Veit Bertram ; Betreuer: Rainer M. Bohle." Saarbrücken : Saarländische Universitäts- und Landesbibliothek, 2019. http://d-nb.info/1197612076/34.

Abdo, Mustafa [Verfasser]. "Prediction of patients’ response to immune checkpoint inhibitors in the treatment of advanced NSCLC / Mustafa Abdo." Göttingen : Niedersächsische Staats- und Universitätsbibliothek Göttingen, 2021. http://d-nb.info/1228623848/34.

Tam, Yee-san Issan. "Epidermal growth factor receptor (EGFR) mutations and phosphorylation pattern in non-small cell lung cancer (NSCLC)." Click to view the E-thesis via HKUTO, 2008. http://sunzi.lib.hku.hk/hkuto/record/B40687673.