Transcriptomic Analysis of Radio-resistant A549 Cells Identify Stemness Gene Signature to Predict Radiotherapy Response in Lung Adenocarcinoma Patients

Transcriptomic Analysis of Radio-resistant A549 Cells Identify Stemness Gene Signature to Predict Radiotherapy Response in Lung Adenocarcinoma Patients | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Transcriptomic Analysis of Radio-resistant A549 Cells Identify Stemness Gene Signature to Predict Radiotherapy Response in Lung Adenocarcinoma Patients Murali MS Balla, Pooja Melwani, Sheri Vidya Rani, Dhruv Das, Vanshikha Gupta, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7240316/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Despite major technical advancements, the prediction of radiotherapy clinical outcome is a major challenge to radiation oncologists due to lack of suitable predictive biomarkers. To address this, a radio-resistant cell line (RR) has been generated from A549 cells (CC) after fractionated doses followed by clonogenic assay. On evaluation these cells showed cancer stem like features. The transcriptomic data of RR versus CC cells were analysed. Out of 658 differentially expressed genes, RNA expression of 60 genes were found to be correlated (p < 0.05) with transcriptomic data of radiotherapy treated lung adenocarcinoma patients obtained from the TCGA database. Binary logistic regression of these 60 selected genes resulted in identification of seven genes (KCNB1, UNC13A, RIMS2, KCNH3, TOX2, SYTL3 and NR3C2) which showed significant (p < 0.05) association with response to radiotherapy. Data was employed to predict radiotherapy response in patients using machine learning algorithms [KNN Cosine]. Algorithms were trained on 80% and tested on 20% of patient’s data. Accuracy was 90% for the model in predicting radiotherapy outcome. When nomogram analysis was performed based on the results of KNN Cosine model, it showed a positive likelihood ratio of 3.45, suggesting potential prognostic nature of this gene signature for radiotherapy outcome in lung cancer patients. Health sciences/Biomarkers Biological sciences/Cancer Biological sciences/Computational biology and bioinformatics Biological sciences/Genetics Health sciences/Oncology Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Introduction Lung carcinoma is the leading cause of cancer death worldwide accounting for 18% of total cancer deaths 1 . Radiotherapy (RT) is one of the main strategies in the treatment of non-small cell lung cancer patients. Major challenges before the radiation oncologists involve tumour recurrence and poor prognosis of the RT response 2 . It was shown that despite having improvement in the combined treatment modalities, recurrence is observed in 30 to 50% of lung adenocarcinoma patients 4 , 5 . Tumor recurrence was mainly attributed to the intra-tumoral heterogeneity developed during the course of radiotherapy 3 . It was later shown that during cancer treatment by fractionated doses of radiation, some of the cells in tumour acquire radio-resistance and these cells were shown to display the properties of stem cells [also termed as cancer stem cells (CSC)] thus resulting in the failure of radiation therapy and occurrence of metastasis 6 , 7 . Despite the several technological advancements in RT techniques, the prediction of clinical outcome is one of the major challenges before the radiation oncologists. Hence, the knowledge about the biomarkers which can predict the RT outcome is of high priority for the clinicians to improve the treatment protocols and minimize the adverse effects of cancer RT. In this regard, several studies showed development of cellular and secretome based biomarkers 8 , 9 . Moreover, gene signatures studied from microarray or transcriptomic analysis in stage I compared with stage III lung adenocarcinoma patients 8 – 14 mainly address the prediction of overall survival and prognosis of patients. The major genes (CASP4, LAMB1, ADM, AKAP12, ARHE, SLC20A1, PDE7A and RECA) reported in these studies 10 – 16 are pertaining to processes of apoptosis, cell adhesion, cell signalling, receptors, enzymes and cellular metabolism pathways for disease progression. Few studies also attempted to evaluate the signatures consisting of 82 genes and PI3K pathway gene signature which predicts therapy recurrence in lung cancer patients 17 , 18 . However, there are only limited studies developing molecular biomarkers for prediction of the outcome of RT in lung cancer patients. A recent study used the ability of six gene signatures (APOBEC3B, GOLM1, FAM117A, KCNQ1OT1, PCDHB2 and USP43) in predicting the RT response in lung cancer patients 19 . In this study, to identify gene signatures non-small cell lung cancer cells of different radio-sensitivity were employed after an acute dose of radiation followed by microarray analysis. However, it was previously shown that the response of radiation is different in case of acute than fractionated doses. Gene expression profile of prostate cancer cells has been found to be different in case of acute dose of 10 Gy than multi-fractionated doses (10 x 1 Gy) 20 .In clinical practice, fractionated doses of radiation delivered during RT results in killing of the sensitive population of cells, however, the surviving fraction of cells acquire radio-resistance during the course of treatment. This population of cancer cells, which is known to be enriched with cancer stem cells (CSCs), not only exhibit radio-resistance but also result in chemo-resistance and metastasis. Hence, cancer cells derived from radio-resistant cells survived after fractionated doses of radiation provide better cell models resembling to clinical RT conditions. Moreover, the gene signature pertaining to stemness in radio-resistant cancer cells would provide better predictive biomarker for the clinical outcome of RT, which however not well reported in the literature for the lung cancer. Transcriptomic studies using patient samples are limited with heterogeneity in cell types and availability of samples after RT. Identification of gene signatures using in vitro models and its correlation at the patient level would not only facilitate the development of suitable biomarkers for RT response but also aid in evaluation of molecular targets with ease for developing suitable inhibitors/drugs for bench-to-bedside translation of the scientific knowledge. With this background, radio-resistant cells from A549 cells were generated by exposing to multiple fractions of doses (5 Gy) followed by clonogenic assay and after each fraction of radiation dose, surviving cells were treated for subsequent doses of radiation. Further transcriptomic profiles of radio-resistant cells exhibiting CSC features were utilized to identify gene signature for predicting RT response in lung adenocarcinoma patients selected from TCGA database. Materials and Methods Cell Culture A549 lung adenocarcinoma cells were obtained from National Centre for Cell Sciences, Pune, India. These cells were cultured in DMEM supplemented with 10% fetal bovine serum (FBS) (Hi-media, Mumbai, India), antibiotics [100 U/mL Penicillin and 100 mg/mL Streptomycin (Sigma)] and maintained at 5% CO 2 at 37°C in humidified incubator. Generation of radio-resistant (RR)cells and clonogenic assay : RR A549 cells were generated by clonogenic assay. After 80% confluency, cells were seeded at a density of 500 cells per 100 mm culture dish for overnight and then were either sham-irradiated or exposed to 5Gy of γ-radiation using Bhabhatron Cobalt-60 Teletherapy Unit (Panacea Medical Technologies, Bangalore; dose rate: 1 Gy/min). Control and irradiated cells were cultured for 10–14 days. Colonies produced by surviving cells were trypsinized and again seeded for next dose of irradiation (as mentioned in this section). This procedure was continued for 4 and 8 doses of each 5 Gy until a cumulative dose of 20Gy (4 x 5 Gy) (R) and 40Gy (8 x 5 Gy) (RR) was reached. Sham treatment to parental A549 cells was maintained for equal passage number as that of irradiated A549 cells.These cells were used as control and referred as CC. The stocks of CC, R and RR were cryopreserved to be used for future experiments. Where ever required, clonogenic assay was performed to calculate survival fraction for CC, R and RR cells as mentioned previously 21 . Single Cell Assay (SCA) SCA of CC and RR cells were carried out as described previously 21 , 22 . Briefly, after harvesting of the CC and RR cells by trypsinisation, cells were counted and serially diluted in complete culture medium (1000 cells/mL). In each well of 96-well plate, 100 µL of cell suspension was dispensed to obtain single cell per well. Wells were randomly checked (at least 20 wells) under a microscope to confirm single cell in each well. Culture plates were incubated for 10–12 days in culture conditions supplemented with 10% FBS and antibiotics. Percent clone forming ability (CFA) was calculated as follows CFA= (Number of colonies formed /Number of cells plated) X100 Immunocytochemistry : Cells obtained from CC and RR cultures were placed on glass cover slip for overnight. From these cultures medium was removed, cells were washed with PBS and fixed using ice-cold acetone for 10 min. The fixed cells were washed three times with PBS. Cells were incubated at room temperature for 30 min with PBS containing 1% BSA followed by 30 min incubation with CD44-FITC (dilution: 1:100; BD, Pharmingen, USA) and CD24-PE antibodies (dilution: 1:200; BD, Pharmingen, USA). Subsequently, cells were washed with PBS and slides were mounted with anti-fade reagent (Molecular Probes, USA) for observing under fluorescence microscope (Nikon, Japan).Where ever required, for γH2AX immunocytochemistry, CC and RR cells were seeded on clover slips for overnight. Cells were irradiated with 2 Gy followed by incubation in culture conditions. Cells were fixed at 3, 6 and 18 h after irradiation to evaluate the double strand breaks using γH2AX –FITC immunocytochemistry mentioned previously 21 , 22 . Semi-quantitative reverse transcriptase PCR Total RNA was isolated from CCand RR cells using TRI reagent (Sigma, USA) and quantified total RNA by Nanodrop 2000 (Thermo Scientific, USA). Two microgram of total RNA was reverse transcribed using Superscript Vilo cDNA Synthesis Kit (Life Technologies, USA). Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as an internal control and reaction without template as a negative control. Sequence of primers for human OCT4, NANOG, CD166, Sox2, TERT, CCSP, GDC, NR3C2, GAPDH and β-actin are mentioned in Supplementary Table S1 with respective product sizes. Intensity of each band was normalised with control genes and compared as fold change with respective genes. Real-Time PCR: Real-time PCR was performed using SYBRgreen 2X mix (Quiagen, USA) in triplicates and quantitation was performed in Qiagen Rotor- Gene Q system and analysed in Microsoft Excel 2007. The expression of GAPDH gene was the internal control and differential expression was determined by the standard formulafor fold change = 2 − δδCt WhereδδCt = δCt treated-δCt untreated and δCt = Ct Gene-Ct GAPDH Unpaired t test was used to get the p-value. The sequence of primers used in this study is given in Supplementary Table S1 . Western Blot Cell lysates of the samples were collected using 1xSDS sample buffer and proteins were seperated in an 8–12% SDS-PAGE. The resolved proteins were transferred using a semi-dry transfer apparatus (Bio-rad, USA) onto nitrocellulose membranes. To visualize proteins 5% bovine serum albumin was used as a blocking solution. The membranes were incubated with Oct4 (Cat#MA514845; Invitrogen, USA), NANOG (Cat#MA1-017; Invitrogen, USA) and β-tubulin (Cell Signalling Technologies; USA) antibodies diluted in 1xTBST containing 5% blocking solution for overnight at 4ºC on an orbital shaker. Subsequently, membranes were washed three times using 1xTBST and then incubated in secondary antibody for 1 hr. After three washes, freshly made ECL solution from Bio-Rad USA was used to develop the signal generated from HRP. Library preparation and sequencing Total RNA was extracted from two experimental replicates of cells obtained from CC and RR cells using RNeasy mini kit (Qiagen, Hilden, Germany). RNA concentration and integrity were determined with Nanodrop 1000 (Nanodrop, Wilmington, DE, USA). Libraries were prepared using the Truseq RNA sample prep kit V2 as per the manufacturer’s protocol (Illumina, San Diego, CA, USA). Briefly, ploy (A) RNA was enriched from total RNA samples using oligodT- attached beads. Double strand cDNA was prepared from enriched RNA samples. cDNA from each sample was attached to a specific index adapter, and the cDNA library was further amplified using index adapters. All the libraries were validated using Agilent Technologies 2100 Bioanalyzer, normalized, and pooled together. 100 bp single-end sequencing was performed on Illumina HiSeq 2500 system (CDAC, NCCS, Bangalore) to get a minimum of 10 million reads per library. RNAseq data from TCGA database of lung cancer patients : Lung adenocarcinoma patients (n = 50, 23 males and 27 females) treated with RT were considered for comparison with in vitro NGS data and predictive analysis using machine learning tools. For this, patient’s RNA Seq files were downloaded from repository of free access from Genomic Data Commons at National Cancer Institute data portal ( https://portal.gdc.cancer.gov/repository ) 23 . TCGA ID numbers and other details of selected patients used in the present study were provided in Supplementary Table S2 and other information about the data is available on ( https://portal.gdc.cancer.gov/repository ). Lung cancer patients were divided into RT responders (n = 21) and non-responders (n = 29) based on tumour absence or presence after treatment. When transcriptomics data of patients were evaluated, the DESeq2 plots of all the samples were not separately clustered based on responders versus non-responders (Supplementary Figure S1 ). In order to reduce the noise from the samples with overlapping variance, samples were selected based on high variance (distant clusters) in the PCA plots of responders and non-responders with respective controls (Supplementary Figure S2 ). To evaluate the genes which have potential to predict RT response, genes wereselected based on significant correlation (Pearson’s correlation) p < 0.05 with the in-vitro data. These selected patients RNA seq files were used to get the DEGs and thensignificantly correlated genes were short-listed with respect to in-vitro RR DEGs data. Binary logistic regression was performed on variance stabilizing transformation(VST) normalised values of individual genes for the outcome of RT response. Further the VST normalised values of selected genes from all the patients (non-responders (n = 29) versus responders (n = 21) to RT) were used to perform predictive analysis (machine learning classification models [Cosine KNN]) using Matlab 2019b (MathworksInc., USA)(Fig. 1 ). Bioinformatics analysisfor in vitro and patient’s data obtained from TCGA database :The quality of RR-CC transcriptomic data was checked using FASTQC. Sequence reads were mapped to the human reference genome GRCh38.106 (hg19) using splice aware aligner RNA STAR Galaxy Version 2.7.8a. The number of raw reads and number of reads were successfully aligned for each of the samples. Gene counts from the aligned BAM files were estimated using feature counts. DESeq2 was run on the counts data of RR versus CC cells. Raw data files have been submitted GEO website database ( https://www.ncbi.nlm.nih.gov/geo ) and the accession number is GSE221796. Genes having Log2FC |1.3| and p-adj < 0.05 were selected for further analysis. Volcano plot and heat map were generated using volcano plot (Galaxy Version 0.0.5) and Heatmap2 (Galaxy Version 3.1.1 + galaxy1), respectively, which can be used by free access ( www.usegalaxy.org ) 24 . Further the patient’sclinical details were obtained from TCGA repository as mentioned in previous Section. Respective count files (rnaseq.augmentedstargenecounts.tsv) were downloaded for each patient. Total number of patients selected for the study were RT responders (n = 21; controls n = 4) and non-responders (n = 29; controls n = 4). Respective Controls represents samples from non-tumour cut boundary samples from the lung tissue. Machine Learning Algorithms To validate the role of identified genes (from in-vitro radio-resistant cells) in predicting radiation response in lung cancer patients, we applied classification learner app (Discriminant analysis, Naive Bayes, support vector machine etc.) in Matlab 2019b. The dataset was randomly split into 80% training set and 20% testing set cohorts. For all models, hyper-parameter tuning was performed using grid search with complete dataset for a given signature. Models were trained with identified parameters on training cohort using 10-fold cross-validation with five repeats. The trained models were tested on test cohort and respective AUC was plotted. The model having better fit, high AUC and accuracy was selected for further testing and nomogram analysis. A workflow of analysis and machine learning is shown in Fig. 1 . Statistical analysis Mean values are out of three independent experiments and error bars are standard deviation. Statistical analysis was performed using Origin 8 software. Student’s t-test was used to compare the means. Unless mentioned, values were considered significantly different at p < 0.05. Results Radio-resistant A549 cell line showed cancer stem-like nature: To identify gene signatures of radio-resistant lung cancer cells, we first generated the radio-resistant (RR) cells, which were evaluated for their radio-resistance by clonogenic assay after treatment with increasing doses (2-10 Gy) of radiation (Figure 2a,b).We found higher survival of RR cells after γ-irradiation, whereas the CC and R cells showed lower clonogenic survival. Since RR cells showed highest radio-resistance, further experiments were conducted using RR cells. Additionally, the radio-resistant nature of RR cells was also evaluated by the measurement of magnitude of radiation-induced DNA double strand break using γH2AX. In response to γ-irradiation, % of γH2AX positive cells was significantly lower in RR cells than CC cells at 3, 6 and 18 h of post-irradiation (Figure 2c) suggesting radio-resistant nature of RR cells. Presence of cancer stem cells is one of the well-known variables to determine the magnitude of radio-resistance 6, 7 . As RR cells were resistant to γ-irradiation, the expression of pluripotent genes (Oct4, Nanog, Sox2, TERT and CD166) were examined in CC and RR cells. Results showed significantly higher level of expression of these genes in RR cells than CC cells (Figure 3 a and b).Out of genes studied, Oct 4 and TERT showed highest fold change in expression in RR cells. Two of the pluripotent genes (Oct-4 and Nanog) were also evaluated using western blot which further corroborates the RT-PCR data (Supplementary Figure S3a). CSCs are known to form clone from single cell due to their self-renewal ability. Hence, to further validate the stemness feature in RR cells, the clone forming abilitywas evaluated employing single cell assay. We found that clone forming ability of RR cells was significantly higher than CC cells (Figure 3 c). CD44 + CD24 - is one of the hallmark features of CSCs, hence magnitude of CD44 + CD24 - was quantified in RR and CC cells. Our results showed about 3 folds higher fraction of CD44 + CD24 - in cultures of RR cells (~75 %) than CC cells(~25 %) (Figure 3d,e). These set of resultssuggest higher CSC-like features of RR cells, which might be contributing to resistance nature of RR cells. Transcriptomic profile of A549 RR cell lines NGS analysis was performed in RR cells and compared to the respective CC cells as mentioned in materials and methods. Compared to their respective controls, the number of differential expressed genes (DEGs) at FDR 1.3) of RR versus CC showed that 157 genes were up-regulated and 293 were down-regulated (Figure 4a). Gene enrichment analysis was performed using Panther Gene Ontology (GO) and the top 3 up/down-regulated processes with highest enrichment scoresare listed in Figure 4b. Our study showed enrichment of cell adhesion pathways in RR cells than controls suggesting higher stemness in RR cells. Higher enrichment score of blood vessel development pathways in RR cells seems to be associated with increased angiogenic features due to CSCs. A plasticity of cancer cells towards cancer stem cells would result in de-differentiation and thus down-regulation of differentiation processes 25 . A549 cells being epithelial in nature, the down-regulation of fat cell and epithelial cell differentiation processes (enrichment score 7.63 and 3.28, respectively) in RR cells is in line of their higher CSC like features of RR cells as observed in Figure 3 a-d. De-regulatedexpression of cell adhesion integrin subunits has been reported in breast and lung cancer spheroids enriched with CSCs and the adhesion in these cancer cells has been inhibited by antagonists of integrins 26 . It was also shown that up-regulation of COL17A1, a cell-adhesion molecule in colorectal cancer stem cells in organoid model, which become more sensitive to chemotherapy when COL17A1 was knocked out 27 . The volcano plots (Figure 4c) and heat maps (Figure 4d) of RR versus CC cells are also shown to compare and represent the expression profile of genes.The top 15 up- and down-regulated genes in RR versus CC cells were listed in Table 1. Out of the analysis, some of the genes which are up-regulated in cells obtained from RR cells were GRHL3, KRT19 and ALDH1A3. These genes were known to be expressed in cancer stem cells 28, 29, 30 suggesting enrichment of CSC features in RR cells. Further, NGS data in these cell lines were validated by real time PCR for the selected genes ALDH1A3, H19 and GHRL3 in samples of RR cell lines(Figure 3d).Genes like NID2, LGR5, and VCAN were found to be down-regulated. The DEGs among RR cells which were significantly up-regulated in this study were FHL1, LAMA4, CDCP1, ALX1, PCDH7, SAA1 and SUSD2. The genes, which were significantly down-regulated, were TRPA1, KDM5D, TTTY15, CLDN2, ANXA13, RHOBTB1, LIMCH1 and RBP4. Correlation of transcriptomic data of RR cells with radiotherapy treated lung cancer responders and non-responder patients selected from TCGA database To identify the signature genes from the transcriptomic data of RR versus CC groups, genes that are expressed in in-vitro system were correlated with patient data from TCGA database. Analysis showed that 60 genes were significantly correlated with selected patient’s transcriptomic data.To evaluate the above short-listed 60 individual genes to predict the RT outcome in lung cancer patients, binary logistic regression was applied. Out of these 60 genes, 7 genes (KCNB1, UNC13A, RIMS2, KCNH3, TOX2, SYTL3 and NR3C2) showed significant association (p<0.05) in predicting RT response in lung cancer patients (Table 2). One of the seven genes NR3C2 was validated using semi-quantitative RT-PCR to confirm transcriptomic data (Supplementary Figure S3b) Further, analysis was carried out using machine learning algorithms to evaluate the ability of these 7 genes together in predicting the RT outcome. Non-responders (n=29) and responders (n=21) data was divided in to training (80%) and testing (20%) sets followed by analysis using machine learning tools in Matlab 2019b.Application for all the available classification algorithms models was applied to the training data set for the best fit. KNN cosine model has shown higher accuracy (76%) with AUC of 0.80 (Figure 5a).Upon testing of the model, the accuracy was found to be 90% in predicting RT response. To evaluate the predictive ability of these gene signatures, nomogram analysis was performed based on ability of gene signature to predict responders and non-responders to RT. The results of KNN cosine model were used to generate the nomogram. The positive likelihood ratio of 7 gene signature was found to be 3.45 suggesting their predictive ability for RT response in lung cancer patients (Figure 5 b). Discussion To identify genes signature associated with stemness, which can predict RT response in lung adenocarcinoma patients, initially RR cell line was generated by treating clonogenic cells with a dose of 5 Gy/fraction eight times. The stem cell genes like Oct4, Nanog, Sox2, TERT and CD166 were highly up-regulated compared to the parental CC cells. Further these results were corroborated by significantly higher percent clone forming ability and expression of CD44 + CD24 - markers in RR cells suggesting stem like nature of these cells. After confirmation of RR cells for their stem-like nature, cells were evaluated for transcriptomic profiles of CSCs. Analysis of data has shown that up-regulated genes in this study like RIMS2, SFRP5, CCL5, ALDH1A3, H19 and VIPR1 and down-regulated genes like EIF1AY, RARB, HNF4A and SLC27A2 have been shown to be differentially expressed in CSCs of different tumours 31-36 . Furthermore, this study also focussed on transcriptomic data of RR versus CC cells and evaluated for genes which can predict RT response in lung adenocarcinoma patients. Total of 60 genes have been shortlisted, which has significant correlation between in-vitro samples and selected patient’s transcriptomic data. Among these, seven genes have been shortlisted which showed significant association in explaining the RT response by binary logistic regression. Out of these 7 genes, 5 genes (KCNB1, UNC13A, RIMS2, KCNH3 and TOX2) were up-regulated and 2 genes (SYTL3 and NR3C2) were down-regulated. KCNB1 expression was shown to be up-regulated in pancreatic cancer stem cells and blocking this potassium voltage-gated channel with 4-aminopyridine inhibited the proliferation of the cells. This was also proposed to be a potential target for pancreatic cancer 37 . In our study, we have shown that KCNB1 gene is highly up regulated in radio-resistant cell line. Additionally, our study has shown its usefulness as a signature gene to predict RT response in lung adenocarcinoma patients. It was reported earlier that UNC13A was shown to be having single nucleotide polymorphisms in keratinocyte carcinomas and this gene’s role in reprogramming human astrocytes into functional neurons through notch signalling 38, 39 . In our study, UNC13A gene expression is up-regulated in radio-resistant cell line and patients’ samples. The genes RIMS2 and KCNH3 expression was shown to be up-regulated in various other cancers 40-43 . In contrast some groups have shown the down-regulation of RIMS2 in glioma and breast cancers 44, 45 . Additionally, it was also shown RIMS2 is highly up-regulated in mesenchymal stem cells of umbilical cord blood 46 . Our data showed that RIMS2 was up-regulated in transcriptomic data of RR cells and patient samples. Its role in radio-resistance is yet to be evaluated. KCNH3 was shown to be highly expressed in ovarian cancers and cardiac progenitor cells. It was suggested that high expression of this gene might be an independent predictor of poor overall disease free survival 47-49 . Our data showed that KCNH3 gene is highly up regulated in radio-resistant cell line and this gene has potential in differentiating RT responders from non-responders. The gene TOX2 expression was earlier shown to be associated with worst overall survival of natural killer/T cell lymphoma patients 50 . It is also showed that its role in RUNX3-TOX2-Super Enhancer-TOX2-Protein of regenerating liver 3 regulatory pathway of natural killer T cell lymphoma biology. However, its role in radio-resistance and stemness is not clearly established. Here in our study, TOX2 is highly expressed in radio-resistant cell lines of A549 and has ability in differentiating RT response of lung adenocarcinoma patients (Table 2). Furthermore, it would be very interesting to evaluate these genes (KCNB1, UNC13A, RIMS2, KCNH3 and TOX2) role in radiation resistant tumors. NR3C2 and SYTL3 were shown to be down-regulated in different cancers 51, 52 . Our data also showed that NR3C2 and SYTL3 genes were down-regulated in RR versus CC cells and patient samples. SYTL3 expression and its functional evaluation have not been done in cancers but its knockdown is correlated with the neuronal cell migration in embryonic mouse cortex. When all these seven genes were used together to predict the RT response using machine learning algorithms, it showed potential in differentiating RT non-responders from responders. In conclusion, this study brings out valuable information about the genes relevant to CSC like cells and possible role in resistance nature of lung carcinoma cells. Additionally seven genes signature was obtained from correlating in-vitro RR cell line model and patient’s data. In this effort, the predictive ability of the signature genes helps in identifying non-responders and before translating them to clinics it is needed to be evaluated in a separate prospective study in radio-therapy treated lung cancer patients. Finally, the functional evaluation of these genes in the proposed signature needs to be evaluated for their ability to sensitize the RR lung cancer cell models which may have applicability in future RT treatment protocols. Declarations Funding Declaration: Authors acknowledge Bhabha Atomic Research Centre, Department of Atomic Energy, Government of India for financial support of this work. Author Contribution M.M.S.B conducted the experiments, analyzed the results, designed the study, wrote and edited the manuscript, P.K.M helped in analysis of data, writing and editing the manuscript, D.D has analyzed data using machine learning tools in Python, S.V.R and VG conducted part of the experiments in the manuscript, H.S.B, S.M, N.G, N.V.S have helped in conducting the RNA Seq protocols on the samples. B.N. P analysed the results, designed the study and edited the manuscript. Data Availability Raw data files have been submitted GEO website database (https://www.ncbi.nlm.nih.gov/geo) and the accession number is GSE221796 References Sung, H. et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. : 71: 209 – 49.doi: 10.3322/caac.21660. Epub 2021 Feb 4. PMID: 33538338. (2021). Terada, Y. et al. Radiotherapy for local recurrence of non-small-cell lung cancer after lobectomy and lymph node dissection-can local recurrence be radically cured by radiation? Jpn J Clin Oncol. ;50(4):425 – 33. (2020). 10.1093/jjco/hyz188 . PMID: 31926488. Herbst, R. S., Morgensztern, D. & Boshoff, C. The biology and management of non-small cell lung cancer. Nature. ; 553(7689):446 – 54. (2018). 10.1038/nature25183 . PMID: 29364287. Le Chevalier, T. et al. Radiotherapy alone versus combined chemotherapy and radiotherapy in nonresectable non-small-cell lung cancer: First analysis of a randomized trial in 353 patients. J. Natl. Cancer Inst. 83, 417–23. (1991). 10.1093/jnci/83.6.417 . PMID: 1847977. Curran, W. J. Jr. et al. Sequential vs. concurrent chemoradiation for stage III non-small cell lung cancer: Randomized phase III trial RTOG 9410. J. Natl. Cancer Inst. 103 , 1452–1460. 10.1093/jnci/djr325 (2011). Nguyen, L. V., Vanner, R., Dirks, P. & Eaves, C. J. Cancer stem cells: an evolving concept. Nat Rev Cancer. ; 12(2):133 – 43. (2012). 10.1038/nrc3184 . PMID: 22237392. Baccelli, I. & Trumpp, A. The evolving concept of cancer and metastasis stem cells. J. Cell. Biol. 198 (3), 281–293. 10.1083/jcb.201202014 (2012). PMID: 22869594; PMCID: PMC3413352. Sant, G. R., Knopf, K. B. & Albala, D. M. Live-single-cell phenotypic cancer biomarkers-future role in precision oncology? NPJ Precis Oncol. ;1(1):21. (2017). 10.1038/s41698-017-0025-y . PMID: 29872705. Stastna, M. & Van Eyk, J. E. Secreted proteins as a fundamental source for biomarker discovery. Proteomics. ;12(4–5):722 – 35. doi: 10.1002/pmic.201100346. Epub 2012 Jan 19. PMID: 22247067 (2012). Beer, D. G. et al. Gene-expression profiles predict survival of patients with lung adenocarcinoma. Nat. Med. 8 , 816–824. 10.1038/nm733 (2002). Epub 2002 Jul 15. PMID: 12118244. Shedden, K. et al. Gene expression-based survival prediction in lung adenocarcinoma: A multi-site, blinded validation study. Nat. Med. 14 , 822–827. 10.1038/nm.1790 (2008). Epub 2008 Jul 20. PMID: 18641660; PMCID: PMC2667337. Li, Y. et al. Network-based approach identified cell cycle genes as predictor of overall survival in lung adenocarcinoma patients. Lung Cancer . 80 , 91–98 (2013). Epub 2013 Jan 26. PMID: 23357462; PMCID: PMC3595338. Bianchi, F. et al. Survival prediction of stage I lung adenocarcinomas by expression of 10 genes. J. Clin. Investig . 117 , 3436–3444 (2007). PMID: 17948124; PMCID: PMC2030461. Raz, D. J. et al. A multigene assay is prognostic of survival in patients with early-stage lung adenocarcinoma. Clin. Cancer Res. 14, 5565–70. (2008). 10.1158/1078-0432.CCR-08-0544 . PMID: 18765549. Park, Y. Y. et al. Development and validation of a prognostic gene-expression signature for lung adenocarcinoma. PLoS ONE 7, e44225. (2012). 10.1371/journal.pone.0044225 . Epub 2012 Sep 7. PMID: 22970185. Akagi, I. et al. Combination of protein coding and noncoding gene expression as a robust prognostic classifier in stage I lung adenocarcinoma. Cancer Res. 73 , 3821–3832. 10.1158/0008-5472.CAN-13-0031 (2013). Epub 2013 May 2. PMID: 23639940. Tomida, S. et al. Relapse-related molecular signature in lung adenocarcinomas identifies patients with dismal prognosis. J. Clin. Oncol. 27, 2793–99.doi: 10.1200/JCO.2008.19.7053. Epub 2009 May 4. PMID: 19414676. (2009). Larsen, J. E. et al. Gene expression signature predicts recurrence in lung adenocarcinoma. Clin. Cancer Res. 13, 2946–54. (2007). 10.1158/1078-0432.CCR-06-2525 . PMID: 17504995. Peinado-Serrano, J. et al. A Six-Gene Prognostic and Predictive Radiotherapy-Based Signature for Early and Locally Advanced Stages in Non-Small-Cell Lung Cancer. Cancers (Basel) . 14 (9), 2054. 10.3390/cancers14092054 (2022). PMID: 35565183; PMCID: PMC9099638. Eke, I. et al. Radiotherapy alters expression of molecular targets in prostate cancer in a fractionation- and time-dependent manner. Sci Rep. ;12(1):3500. (2022). 10.1038/s41598-022-07394-y . PMID: 35241721. Balla, M. M. S., Yadav, H. D. & Pandey, B. N. Tumorsphere assay provides a better in vitro method for cancer stem-like cells enrichment in A549 lung adenocarcinoma cells. Tissue Cell. 60 , 21–24 (2019). Epub 2019 Jul 16. PMID: 31582014. Balla, M. M., Ningthoujam, R. S., Kumar, M., Bandekar, J. R. & Pandey, B. N. Cellular and spectroscopic characterization of cancer stem cell-like cells derived from A549 lung carcinoma. J Cancer Res Ther. (2016). Jul-Sep;12(3):1144-52 10.4103/0973-1482.171365 . PMID: 28054526. Sanchez-Vega, F. et al. Oncogenic Signaling Pathways in The Cancer Genome Atlas. Cell 173 (2), 321–337e10. 10.1016/j.cell.2018.03.035 (2018). PMID: 29625050; PMCID: PMC6070353. Batut, B., van den Beek, M., Doyle, M. A. & Soranzo, N. RNA-Seq Data Analysis in Galaxy. Methods Mol Biol. ; 2284:367–392. (2021). 10.1007/978-1-0716-1307-8_20 . PMID: 33835453. Vlashi, E. & Pajonk, F. Cancer stem cells, cancer cell plasticity and radiation therapy. Semin Cancer Biol. ;31:28–35. (2015). 10.1016/j.semcancer.2014.07.001 . Epub 2014 Jul 12. PMID: 25025713. Paolillo, M. et al. Stem-Like Cancer Cells in a Dynamic 3D Culture System: A Model to Study Metastatic Cell Adhesion and Anti-Cancer Drugs. Cells. ;8(11):1434. (2019). 10.3390/cells8111434 . PMID: 31766310. Ohta, Y. et al. Cell-matrix interface regulates dormancy in human colon cancer stem cells. Nature 608 (7924), 784–794. 10.1038/s41586-022-05043-y (2022). Epub 2022 Jul 7. PMID: 35798028. Lin, Z. et al. Murine interfollicular epidermal differentiation is gradualistic with GRHL3 controlling progression from stem to transition cell states. Nat Commun. ;11(1):5434. (2020). 10.1038/s41467-020-19234-6 . PMID: 33116143. Sullivan, K. E., Rojas, K., Cerione, R. A., Nakano, I. & Wilson, K. F. The stem cell/cancer stem cell marker ALDH1A3 regulates the expression of the survival factor tissue transglutaminase, in mesenchymal glioma stem cells. Oncotarget 8 (14), 22325–22343. 10.18632/oncotarget.16479 (2017). PMID: 28423611; PMCID: PMC5410226. Lecerf, C., Peperstraete, E., Le Bourhis, X. & Adriaenssens, E. Propagation and Maintenance of Cancer Stem Cells: A Major Influence of the Long Non-Coding RNA H19. Cells 9 (12), 2613. 10.3390/cells9122613 (2020). PMID: 33291403; PMCID: PMC7762009. Secco, M. et al. Gene expression profile of mesenchymal stem cells from paired umbilical cord units: cord is different from blood. Stem Cell Rev Rep. ;5(4):387–401. (2009). 10.1007/s12015-009-9098-5 . PMID: 20058202. Behera, A., Ashraf, R., Srivastava, A. K. & Kumar, S. Bioinformatics analysis and verification of molecular targets in ovarian cancer stem-like cells. Heliyon 6 (9), e04820. 10.1016/j.heliyon.2020.e04820 (2020). PMID: 32984578; PMCID: PMC7492822. Kawakami, K. et al. Secreted frizzled-related protein-5 is epigenetically downregulated and functions as a tumor suppressor in kidney cancer. Int. J. Cancer . 128 (3), 541–550. 10.1002/ijc.25357 (2011). PMID: 20340127; PMCID: PMC3940353. Huang, R. et al. CCL5 derived from tumor-associated macrophages promotes prostate cancer stem cells and metastasis via activating β-catenin/STAT3 signaling. Cell. Death Dis. 11 (4), 234doi. 10.1038/s41419-020-2435-y (2020). PMID: 32300100; PMCID: PMC7162982. Sastry, K. S. et al. Cytoprotective effect of neuropeptides on cancer stem cells: vasoactive intestinal peptide-induced antiapoptotic signaling. Cell. Death Dis. 8 (6), e2844. 10.1038/cddis.2017.226 (2017). PMID: 28569785; PMCID: PMC5520887. Su, J. et al. SLC27A2 induces cisplatin resistance in lung cancer stem cells by negatively regulating Bmi1-ABCG2 signaling. Mol. Carcinog. 55 (11), 1822–1832. 10.1002/mc.22430 (2016). Epub 2015 Oct 29. PMID: 26513225. Shiozaki, A. et al. Roles of voltage–gated potassium channels in the maintenance of pancreatic cancer stem cells. Int. J. Oncol. 59 (4), 76. 10.3892/ijo.2021.5256 (2021). Epub 2021 Aug 20. PMID: 34414448. Sunder-Plassmann, R., Geusau, A., Endler, G., Weninger, W. & Wielscher, M. Identification of Genetic Risk Factors for Keratinocyte Cancer in Immunosuppressed Solid Organ Transplant Recipients: A Case-Control Study. Cancers (Basel). ;15(13):3354. (2023). 10.3390/cancers15133354 . PMID: 37444464. Yin, J. C. et al. Chemical Conversion of Human Fetal Astrocytes into Neurons through Modulation of Multiple Signaling Pathways. Stem Cell Reports. ;12(3):488–501. doi: 10.1016/j.stemcr.2019.01.003. Epub 2019 Feb 7. PMID: 30745031 (2019). Secco, M. et al. Gene expression profile of mesenchymal stem cells from paired umbilical cord units: cord is different from blood. Stem Cell Rev Rep. ;5(4):387–401. (2009). 10.1007/s12015-009-9098-5 . PMID: 20058202. Behera, A., Ashraf, R., Srivastava, A. K. & Kumar, S. Bioinformatics analysis and verification of molecular targets in ovarian cancer stem-like cells. Heliyon 6 (9), e04820. 10.1016/j.heliyon.2020.e04820 (2020). PMID: 32984578; PMCID: PMC7492822. Vezzali, R. et al. The FOXG1/FOXO/SMAD network balances proliferation and differentiation of cortical progenitors and activates Kcnh3 expression in mature neurons. Oncotarget. ;7(25):37436–37455. (2016). 10.18632/oncotarget.9545 . PMID: 27224923. Li, Z. et al. KCNH3 Predicts Poor Prognosis and Promotes Progression in Ovarian Cancer. Onco Targets Ther. ;13:10323–10333. (2020). 10.2147/OTT.S268055 . PMID: 33116612. Wei, B., Wang, L. & Zhao, J. Circular RNA hsa_circ_0005114-miR-142-3p/miR-590-5p-adenomatous polyposis coli protein axis as a potential target for treatment of glioma. Oncol. Lett. 21 (1), 58doi. 10.3892/ol.2020.12320 (2021). Epub 2020 Nov 19. PMID: 33281969; PMCID: PMC7709550. Zhang, L., Liu, Z. & Zhu, J. In silico screening using bulk and single-cell RNA-seq data identifies RIMS2 as a prognostic marker in basal-like breast cancer: A retrospective study. Med. (Baltim). 100 (16), e25414 (2021). PMID: 33879671; PMCID: PMC8078249. Secco, M. et al. Gene expression profile of mesenchymal stem cells from paired umbilical cord units: cord is different from blood. Stem Cell Rev Rep. ;5(4):387–401. (2009). 10.1007/s12015-009-9098-5 . PMID: 20058202. Zhang, Y. et al. Epigenomic Reprogramming of Adult Cardiomyocyte-Derived Cardiac Progenitor Cells. Sci Rep. ;5:17686. (2015). 10.1038/srep17686 . Erratum in: Sci Rep. 2017;7:46907. PMID: 26657817. Vezzali, R. et al. The FOXG1/FOXO/SMAD network balances proliferation and differentiation of cortical progenitors and activates Kcnh3 expression in mature neurons. Oncotarget. ;7(25):37436–37455. (2016). 10.18632/oncotarget.9545 . PMID: 27224923. Li, Z. et al. KCNH3 Predicts Poor Prognosis and Promotes Progression in Ovarian Cancer. Onco Targets Ther. ;13:10323–10333. (2020). 10.2147/OTT.S268055 . PMID: 33116612. Zhou, J. et al. Super-enhancer-driven TOX2 mediates oncogenesis in Natural Killer/T Cell Lymphoma. Mol Cancer. ;22(1):69. (2023). 10.1186/s12943-023-01767-1 . PMID: 37032358. Li, X. et al. Nuclear receptor subfamily 3 group c member 2 (NR3C2) is downregulated due to hypermethylation and plays a tumor-suppressive role in colon cancer. Mol. Cell. Biochem. 477 (11), 2669–2679. 10.1007/s11010-022-04449-6 (2022). Epub 2022 May 23. PMID: 35604518. Dong, X. et al. Transcriptional networks identify synaptotagmin-like 3 as a regulator of cortical neuronal migration during early neurodevelopment. Cell Rep. ; 34(9):108802. (2021). 10.1016/j.celrep.2021.108802 . PMID: 33657377. Tables Table 1 and 2 are available in the Supplementary Files section. Additional Declarations No competing interests reported. Supplementary Files SupplemantaryFigures.docx Tables.pptx SupplementaryFileOriginalGelsandBlots.pptx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7240316","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":511308380,"identity":"0076ee7b-e6f6-4c4f-9c8d-af7524f7e258","order_by":0,"name":"Murali MS Balla","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAzElEQVRIie3OoQrCQBzH8f8x2MqJdWHgK0wM40DZq3jsJRYMf8tZtJ8wfIzlGwemgdWwMoRhMMxoENQia7c1wfvEP78v/AEs6weF4KwVpOp7oD0SggrKwQkRyrTriLwC1eNQTaKNcq53AUFkStiWY7HLm2lWLl22F0AZmh5THNUo10T64M5oCTQ0PRmeaiyemY4HJGeOeoSavxPnQtMeCZM16uDYJJJyQWTqm5NonBT1bVUtpKd124bz2Jh0EOED+P33H047bG9ZlvUvXnFYRdFJp8lQAAAAAElFTkSuQmCC","orcid":"","institution":"Bhabha Atomic Research Centre","correspondingAuthor":true,"prefix":"","firstName":"Murali","middleName":"MS","lastName":"Balla","suffix":""},{"id":511308381,"identity":"8d058c96-e382-472e-8f07-e7e68691cbad","order_by":1,"name":"Pooja Melwani","email":"","orcid":"","institution":"Bhabha Atomic Research Centre","correspondingAuthor":false,"prefix":"","firstName":"Pooja","middleName":"","lastName":"Melwani","suffix":""},{"id":511308382,"identity":"0d12510b-8535-45ac-be4c-cab32ae2134b","order_by":2,"name":"Sheri Vidya Rani","email":"","orcid":"","institution":"Bhabha Atomic Research Centre","correspondingAuthor":false,"prefix":"","firstName":"Sheri","middleName":"Vidya","lastName":"Rani","suffix":""},{"id":511308383,"identity":"e8ce33fe-a3fa-4247-a265-938446a7b63d","order_by":3,"name":"Dhruv Das","email":"","orcid":"","institution":"Homi Bhabha National Institute","correspondingAuthor":false,"prefix":"","firstName":"Dhruv","middleName":"","lastName":"Das","suffix":""},{"id":511308384,"identity":"dd25ad5b-ccee-4ecf-b11e-884cd5698f89","order_by":4,"name":"Vanshikha Gupta","email":"","orcid":"","institution":"DRDE, DRDO","correspondingAuthor":false,"prefix":"","firstName":"Vanshikha","middleName":"","lastName":"Gupta","suffix":""},{"id":511308385,"identity":"1438b255-fd47-4b9f-86d9-5c0b769d8394","order_by":5,"name":"Harish Shrikrishna Bharambe","email":"","orcid":"","institution":"Advanced Centre for Treatment, Research and Education in Cancer","correspondingAuthor":false,"prefix":"","firstName":"Harish","middleName":"Shrikrishna","lastName":"Bharambe","suffix":""},{"id":511308386,"identity":"fd38fe86-61a5-48a8-8dbe-68f61f36096c","order_by":6,"name":"Shalaka Masurkar","email":"","orcid":"","institution":"Advanced Centre for Treatment, Research and Education in Cancer","correspondingAuthor":false,"prefix":"","firstName":"Shalaka","middleName":"","lastName":"Masurkar","suffix":""},{"id":511308387,"identity":"1cb00def-952f-49f5-9184-4bb89550bc06","order_by":7,"name":"Nikhil Gadewal","email":"","orcid":"","institution":"Advanced Centre for Treatment, Research and Education in Cancer","correspondingAuthor":false,"prefix":"","firstName":"Nikhil","middleName":"","lastName":"Gadewal","suffix":""},{"id":511308388,"identity":"4a32a705-e0ac-4a12-a6b6-e811f75440bc","order_by":8,"name":"Neelam Vishwanath Shirsat","email":"","orcid":"","institution":"Advanced Centre for Treatment, Research and Education in Cancer","correspondingAuthor":false,"prefix":"","firstName":"Neelam","middleName":"Vishwanath","lastName":"Shirsat","suffix":""},{"id":511308389,"identity":"b186ec5a-2a01-4c53-9a65-f2717d306af1","order_by":9,"name":"Badri Narain Pandey","email":"","orcid":"","institution":"Bhabha Atomic Research Centre","correspondingAuthor":false,"prefix":"","firstName":"Badri","middleName":"Narain","lastName":"Pandey","suffix":""}],"badges":[],"createdAt":"2025-07-29 07:38:18","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7240316/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7240316/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":90929569,"identity":"7bcdd51e-b849-4ac1-ba1a-c991f953aa87","added_by":"auto","created_at":"2025-09-09 16:05:24","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":92836,"visible":true,"origin":"","legend":"\u003cp\u003eWork Flow for the identification of gene signature to predict RT response\u003c/p\u003e","description":"","filename":"Slide1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7240316/v1/34b72db7370b2f8de1133bec.jpg"},{"id":90930753,"identity":"f3fd6ea1-9484-4cfb-b540-64f34efbba5a","added_by":"auto","created_at":"2025-09-09 16:13:24","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":85360,"visible":true,"origin":"","legend":"\u003cp\u003e(a) Clonogenic assay of CC, R and RR cells after irradiation at 0, 4, 6, 8 and 10 Gy (b) Survival fraction of CC, R and RR cells(c) γH2AX positive cells in CC and RR cells at 3, 6 and 18 h after 2 Gy irradiation.\u003c/p\u003e","description":"","filename":"Slide2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7240316/v1/95c814d4f412b4420053e63a.jpg"},{"id":90929572,"identity":"5f1ab00a-9563-4407-8cd3-c79951d374c1","added_by":"auto","created_at":"2025-09-09 16:05:24","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":108081,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003e(\u003c/strong\u003ea, b) Changes in expression of pluripotent and lung cells specific genes of CC and RR cells by semi quantitative RT-PCR and its representative gel image (c) Percent clone forming ability of CC and RR cells obtained from single cell assay (d) Staining of CD44-FITC and CD24-PE in cultures and single cell clones of CC and RR cells counterstained with DAPI for nucleus. Images shown are acquires at 20X magnification, scale bar: 20 µm (e) Percent CD44+ and CD24- cells in A549 cells.\u003c/p\u003e","description":"","filename":"Slide3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7240316/v1/f5f45c0dd2a72d1692e6ecc9.jpg"},{"id":90931606,"identity":"4e0d27a0-fcd5-41dd-ad91-9820998db2e3","added_by":"auto","created_at":"2025-09-09 16:21:24","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":104155,"visible":true,"origin":"","legend":"\u003cp\u003e(a) Transcriptomic analysis of RR versus CC (b) PANTHER overrepresentation using binomial test to determine important pathway genes (c) Volcano plots of top 50 up and down-regulated genes (d) Heat map of top differentially regulated genes (e) Validation of genes (ALDH1A3, H19 and GHRL3) by real-time PCR in RR versus CC cells.\u003c/p\u003e","description":"","filename":"Slide4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7240316/v1/630ab84f90bc287aa196508a.jpg"},{"id":90929570,"identity":"aeb862a9-0e73-4f33-a1f1-8767a603dc9a","added_by":"auto","created_at":"2025-09-09 16:05:24","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":62646,"visible":true,"origin":"","legend":"\u003cp\u003e(a) ROC curve for seven gene signature (KCNB1, UNC13A, RIMS2, KCNH3, TOX2, SYTL3 and NR3C2) (b) Nomogram of 7 genes signature.\u003c/p\u003e","description":"","filename":"Slide5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7240316/v1/1c576b37695e0dbf3b1f3753.jpg"},{"id":92285299,"identity":"20fc3245-abd1-4cc2-a69a-a8058e4ca9c4","added_by":"auto","created_at":"2025-09-26 18:23:36","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1280997,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7240316/v1/5703806e-a338-4b15-a75a-3eb2bd8106e9.pdf"},{"id":90929579,"identity":"47c14592-adb3-4320-9c95-5d7853d2f2f6","added_by":"auto","created_at":"2025-09-09 16:05:24","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":637031,"visible":true,"origin":"","legend":"","description":"","filename":"SupplemantaryFigures.docx","url":"https://assets-eu.researchsquare.com/files/rs-7240316/v1/54721dc5188af3ff2e8a54b5.docx"},{"id":90929575,"identity":"277bc94a-87da-4040-9795-2908ecd09416","added_by":"auto","created_at":"2025-09-09 16:05:24","extension":"pptx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":96077,"visible":true,"origin":"","legend":"","description":"","filename":"Tables.pptx","url":"https://assets-eu.researchsquare.com/files/rs-7240316/v1/98b4cec2948e8f3408d635f1.pptx"},{"id":90929581,"identity":"bbd1f5f6-f43e-4683-b411-51cbb8d4e14a","added_by":"auto","created_at":"2025-09-09 16:05:24","extension":"pptx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":3438539,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryFileOriginalGelsandBlots.pptx","url":"https://assets-eu.researchsquare.com/files/rs-7240316/v1/6a5fa000ae1e5699281053de.pptx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Transcriptomic Analysis of Radio-resistant A549 Cells Identify Stemness Gene Signature to Predict Radiotherapy Response in Lung Adenocarcinoma Patients","fulltext":[{"header":"Introduction","content":"\u003cp\u003eLung carcinoma is the leading cause of cancer death worldwide accounting for 18% of total cancer deaths\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e. Radiotherapy (RT) is one of the main strategies in the treatment of non-small cell lung cancer patients. Major challenges before the radiation oncologists involve tumour recurrence and poor prognosis of the RT response\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e. It was shown that despite having improvement in the combined treatment modalities, recurrence is observed in 30 to 50% of lung adenocarcinoma patients\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e. Tumor recurrence was mainly attributed to the intra-tumoral heterogeneity developed during the course of radiotherapy\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e. It was later shown that during cancer treatment by fractionated doses of radiation, some of the cells in tumour acquire radio-resistance and these cells were shown to display the properties of stem cells [also termed as cancer stem cells (CSC)] thus resulting in the failure of radiation therapy and occurrence of metastasis\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e. Despite the several technological advancements in RT techniques, the prediction of clinical outcome is one of the major challenges before the radiation oncologists. Hence, the knowledge about the biomarkers which can predict the RT outcome is of high priority for the clinicians to improve the treatment protocols and minimize the adverse effects of cancer RT.\u003c/p\u003e\u003cp\u003eIn this regard, several studies showed development of cellular and secretome based biomarkers\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e. Moreover, gene signatures studied from microarray or transcriptomic analysis in stage I compared with stage III lung adenocarcinoma patients\u003csup\u003e\u003cspan additionalcitationids=\"CR9 CR10 CR11 CR12 CR13\" citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003emainly address the prediction of overall survival and prognosis of patients. The major genes (CASP4, LAMB1, ADM, AKAP12, ARHE, SLC20A1, PDE7A and RECA) reported in these studies\u003csup\u003e\u003cspan additionalcitationids=\"CR11 CR12 CR13 CR14 CR15\" citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e are pertaining to processes of apoptosis, cell adhesion, cell signalling, receptors, enzymes and cellular metabolism pathways for disease progression. Few studies also attempted to evaluate the signatures consisting of 82 genes and PI3K pathway gene signature which predicts therapy recurrence in lung cancer patients\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e. However, there are only limited studies developing molecular biomarkers for prediction of the outcome of RT in lung cancer patients. A recent study used the ability of six gene signatures (APOBEC3B, GOLM1, FAM117A, KCNQ1OT1, PCDHB2 and USP43) in predicting the RT response in lung cancer patients \u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e. In this study, to identify gene signatures non-small cell lung cancer cells of different radio-sensitivity were employed after an acute dose of radiation followed by microarray analysis. However, it was previously shown that the response of radiation is different in case of acute than fractionated doses. Gene expression profile of prostate cancer cells has been found to be different in case of acute dose of 10 Gy than multi-fractionated doses (10 x 1 Gy) \u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e.In clinical practice, fractionated doses of radiation delivered during RT results in killing of the sensitive population of cells, however, the surviving fraction of cells acquire radio-resistance during the course of treatment. This population of cancer cells, which is known to be enriched with cancer stem cells (CSCs), not only exhibit radio-resistance but also result in chemo-resistance and metastasis. Hence, cancer cells derived from radio-resistant cells survived after fractionated doses of radiation provide better cell models resembling to clinical RT conditions. Moreover, the gene signature pertaining to stemness in radio-resistant cancer cells would provide better predictive biomarker for the clinical outcome of RT, which however not well reported in the literature for the lung cancer. Transcriptomic studies using patient samples are limited with heterogeneity in cell types and availability of samples after RT. Identification of gene signatures using in vitro models and its correlation at the patient level would not only facilitate the development of suitable biomarkers for RT response but also aid in evaluation of molecular targets with ease for developing suitable inhibitors/drugs for bench-to-bedside translation of the scientific knowledge.\u003c/p\u003e\u003cp\u003eWith this background, radio-resistant cells from A549 cells were generated by exposing to multiple fractions of doses (5 Gy) followed by clonogenic assay and after each fraction of radiation dose, surviving cells were treated for subsequent doses of radiation. Further transcriptomic profiles of radio-resistant cells exhibiting CSC features were utilized to identify gene signature for predicting RT response in lung adenocarcinoma patients selected from TCGA database.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cp\u003e\u003cstrong\u003eCell Culture\u003c/strong\u003e\u003cp\u003eA549 lung adenocarcinoma cells were obtained from National Centre for Cell Sciences, Pune, India. These cells were cultured in DMEM supplemented with 10% fetal bovine serum (FBS) (Hi-media, Mumbai, India), antibiotics [100 U/mL Penicillin and 100 mg/mL Streptomycin (Sigma)] and maintained at 5% CO\u003csub\u003e2\u003c/sub\u003e at 37\u0026deg;C in humidified incubator.\u003c/p\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eGeneration of radio-resistant (RR)cells and clonogenic assay\u003c/b\u003e: RR A549 cells were generated by clonogenic assay. After 80% confluency, cells were seeded at a density of 500 cells per 100 mm culture dish for overnight and then were either sham-irradiated or exposed to 5Gy of γ-radiation using Bhabhatron Cobalt-60 Teletherapy Unit (Panacea Medical Technologies, Bangalore; dose rate: 1 Gy/min). Control and irradiated cells were cultured for 10\u0026ndash;14 days. Colonies produced by surviving cells were trypsinized and again seeded for next dose of irradiation (as mentioned in this section). This procedure was continued for 4 and 8 doses of each 5 Gy until a cumulative dose of 20Gy (4 x 5 Gy) (R) and 40Gy (8 x 5 Gy) (RR) was reached. Sham treatment to parental A549 cells was maintained for equal passage number as that of irradiated A549 cells.These cells were used as control and referred as CC. The stocks of CC, R and RR were cryopreserved to be used for future experiments. Where ever required, clonogenic assay was performed to calculate survival fraction for CC, R and RR cells as mentioned previously\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eSingle Cell Assay (SCA)\u003c/strong\u003e\u003cp\u003eSCA of CC and RR cells were carried out as described previously\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e. Briefly, after harvesting of the CC and RR cells by trypsinisation, cells were counted and serially diluted in complete culture medium (1000 cells/mL). In each well of 96-well plate, 100 \u0026micro;L of cell suspension was dispensed to obtain single cell per well. Wells were randomly checked (at least 20 wells) under a microscope to confirm single cell in each well. Culture plates were incubated for 10\u0026ndash;12 days in culture conditions supplemented with 10% FBS and antibiotics. Percent clone forming ability (CFA) was calculated as follows\u003c/p\u003e\u003c/p\u003e\u003cp\u003eCFA= (Number of colonies formed /Number of cells plated) X100\u003c/p\u003e\u003cp\u003e\u003cb\u003eImmunocytochemistry\u003c/b\u003e: Cells obtained from CC and RR cultures were placed on glass cover slip for overnight. From these cultures medium was removed, cells were washed with PBS and fixed using ice-cold acetone for 10 min. The fixed cells were washed three times with PBS. Cells were incubated at room temperature for 30 min with PBS containing 1% BSA followed by 30 min incubation with CD44-FITC (dilution: 1:100; BD, Pharmingen, USA) and CD24-PE antibodies (dilution: 1:200; BD, Pharmingen, USA). Subsequently, cells were washed with PBS and slides were mounted with anti-fade reagent (Molecular Probes, USA) for observing under fluorescence microscope (Nikon, Japan).Where ever required, for γH2AX immunocytochemistry, CC and RR cells were seeded on clover slips for overnight. Cells were irradiated with 2 Gy followed by incubation in culture conditions. Cells were fixed at 3, 6 and 18 h after irradiation to evaluate the double strand breaks using γH2AX \u0026ndash;FITC immunocytochemistry mentioned previously\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eSemi-quantitative reverse transcriptase PCR\u003c/strong\u003e\u003cp\u003eTotal RNA was isolated from CCand RR cells using TRI reagent (Sigma, USA) and quantified total RNA by Nanodrop 2000 (Thermo Scientific, USA). Two microgram of total RNA was reverse transcribed using Superscript Vilo cDNA Synthesis Kit (Life Technologies, USA). Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as an internal control and reaction without template as a negative control. Sequence of primers for human OCT4, NANOG, CD166, Sox2, TERT, CCSP, GDC, NR3C2, GAPDH and β-actin are mentioned in Supplementary Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e with respective product sizes. Intensity of each band was normalised with control genes and compared as fold change with respective genes.\u003c/p\u003e\u003c/p\u003e\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003eReal-Time PCR:\u003c/h2\u003e\u003cp\u003eReal-time PCR was performed using SYBRgreen 2X mix (Quiagen, USA) in triplicates and quantitation was performed in Qiagen Rotor- Gene Q system and analysed in Microsoft Excel 2007. The expression of GAPDH gene was the internal control and differential expression was determined by the standard formulafor fold change\u0026thinsp;=\u0026thinsp;2\u003csup\u003e\u0026minus;\u0026thinsp;δδCt\u003c/sup\u003e\u003c/p\u003e\u003cp\u003eWhereδδCt\u0026thinsp;=\u0026thinsp;δCt treated-δCt untreated and δCt\u0026thinsp;=\u0026thinsp;Ct Gene-Ct GAPDH\u003c/p\u003e\u003cp\u003eUnpaired t test was used to get the p-value. The sequence of primers used in this study is given in Supplementary Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eWestern Blot\u003c/strong\u003e\u003cp\u003eCell lysates of the samples were collected using 1xSDS sample buffer and proteins were seperated in an 8\u0026ndash;12% SDS-PAGE. The resolved proteins were transferred using a semi-dry transfer apparatus (Bio-rad, USA) onto nitrocellulose membranes. To visualize proteins 5% bovine serum albumin was used as a blocking solution. The membranes were incubated with Oct4 (Cat#MA514845; Invitrogen, USA), NANOG (Cat#MA1-017; Invitrogen, USA) and β-tubulin (Cell Signalling Technologies; USA) antibodies diluted in 1xTBST containing 5% blocking solution for overnight at 4\u0026ordm;C on an orbital shaker. Subsequently, membranes were washed three times using 1xTBST and then incubated in secondary antibody for 1 hr. After three washes, freshly made ECL solution from Bio-Rad USA was used to develop the signal generated from HRP.\u003c/p\u003e\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eLibrary preparation and sequencing\u003c/strong\u003e\u003cp\u003eTotal RNA was extracted from two experimental replicates of cells obtained from CC and RR cells using RNeasy mini kit (Qiagen, Hilden, Germany). RNA concentration and integrity were determined with Nanodrop 1000 (Nanodrop, Wilmington, DE, USA). Libraries were prepared using the Truseq RNA sample prep kit V2 as per the manufacturer\u0026rsquo;s protocol (Illumina, San Diego, CA, USA). Briefly, ploy (A) RNA was enriched from total RNA samples using oligodT- attached beads. Double strand cDNA was prepared from enriched RNA samples. cDNA from each sample was attached to a specific index adapter, and the cDNA library was further amplified using index adapters. All the libraries were validated using Agilent Technologies 2100 Bioanalyzer, normalized, and pooled together. 100 bp single-end sequencing was performed on Illumina HiSeq 2500 system (CDAC, NCCS, Bangalore) to get a minimum of 10\u0026nbsp;million reads per library.\u003c/p\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eRNAseq data from TCGA database of lung cancer patients\u003c/b\u003e: Lung adenocarcinoma patients (n\u0026thinsp;=\u0026thinsp;50, 23 males and 27 females) treated with RT were considered for comparison with in vitro NGS data and predictive analysis using machine learning tools. For this, patient\u0026rsquo;s RNA Seq files were downloaded from repository of free access from Genomic Data Commons at National Cancer Institute data portal (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://portal.gdc.cancer.gov/repository\u003c/span\u003e\u003cspan address=\"https://portal.gdc.cancer.gov/repository\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) \u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e. TCGA ID numbers and other details of selected patients used in the present study were provided in Supplementary Table \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e and other information about the data is available on (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://portal.gdc.cancer.gov/repository\u003c/span\u003e\u003cspan address=\"https://portal.gdc.cancer.gov/repository\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Lung cancer patients were divided into RT responders (n\u0026thinsp;=\u0026thinsp;21) and non-responders (n\u0026thinsp;=\u0026thinsp;29) based on tumour absence or presence after treatment. When transcriptomics data of patients were evaluated, the DESeq2 plots of all the samples were not separately clustered based on responders versus non-responders (Supplementary Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). In order to reduce the noise from the samples with overlapping variance, samples were selected based on high variance (distant clusters) in the PCA plots of responders and non-responders with respective controls (Supplementary Figure \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e). To evaluate the genes which have potential to predict RT response, genes wereselected based on significant correlation (Pearson\u0026rsquo;s correlation) p\u0026thinsp;\u0026lt;\u0026thinsp;0.05 with the in-vitro data. These selected patients RNA seq files were used to get the DEGs and thensignificantly correlated genes were short-listed with respect to in-vitro RR DEGs data. Binary logistic regression was performed on variance stabilizing transformation(VST) normalised values of individual genes for the outcome of RT response. Further the VST normalised values of selected genes from all the patients (non-responders (n\u0026thinsp;=\u0026thinsp;29) versus responders (n\u0026thinsp;=\u0026thinsp;21) to RT) were used to perform predictive analysis (machine learning classification models [Cosine KNN]) using Matlab 2019b (MathworksInc., USA)(Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eBioinformatics analysisfor in vitro and patient\u0026rsquo;s data obtained from TCGA database\u003c/b\u003e:The quality of RR-CC transcriptomic data was checked using FASTQC. Sequence reads were mapped to the human reference genome GRCh38.106 (hg19) using splice aware aligner RNA STAR Galaxy Version 2.7.8a. The number of raw reads and number of reads were successfully aligned for each of the samples. Gene counts from the aligned BAM files were estimated using feature counts. DESeq2 was run on the counts data of RR versus CC cells. Raw data files have been submitted GEO website database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.ncbi.nlm.nih.gov/geo\u003c/span\u003e\u003cspan address=\"https://www.ncbi.nlm.nih.gov/geo\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) and the accession number is GSE221796. Genes having Log2FC |1.3| and p-adj\u0026thinsp;\u0026lt;\u0026thinsp;0.05 were selected for further analysis. Volcano plot and heat map were generated using volcano plot (Galaxy Version 0.0.5) and Heatmap2 (Galaxy Version 3.1.1\u0026thinsp;+\u0026thinsp;galaxy1), respectively, which can be used by free access (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e\u003ca href=\"https://portal.gdc.cancer.gov/repository\" target=\"_blank\"\u003ewww.usegalaxy.org\u003c/a\u003e\u003c/span\u003e\u003cspan address=\"http://www.usegalaxy.org\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) \u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e. Further the patient\u0026rsquo;sclinical details were obtained from TCGA repository as mentioned in previous Section. Respective count files (rnaseq.augmentedstargenecounts.tsv) were downloaded for each patient. Total number of patients selected for the study were RT responders (n\u0026thinsp;=\u0026thinsp;21; controls n\u0026thinsp;=\u0026thinsp;4) and non-responders (n\u0026thinsp;=\u0026thinsp;29; controls n\u0026thinsp;=\u0026thinsp;4). Respective Controls represents samples from non-tumour cut boundary samples from the lung tissue.\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eMachine Learning Algorithms\u003c/strong\u003e\u003cp\u003eTo validate the role of identified genes (from in-vitro radio-resistant cells) in predicting radiation response in lung cancer patients, we applied classification learner app (Discriminant analysis, Naive Bayes, support vector machine etc.) in Matlab 2019b. The dataset was randomly split into 80% training set and 20% testing set cohorts. For all models, hyper-parameter tuning was performed using grid search with complete dataset for a given signature. Models were trained with identified parameters on training cohort using 10-fold cross-validation with five repeats. The trained models were tested on test cohort and respective AUC was plotted. The model having better fit, high AUC and accuracy was selected for further testing and nomogram analysis. A workflow of analysis and machine learning is shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eStatistical analysis\u003c/strong\u003e\u003cp\u003eMean values are out of three independent experiments and error bars are standard deviation. Statistical analysis was performed using Origin 8 software. Student\u0026rsquo;s t-test was used to compare the means. Unless mentioned, values were considered significantly different at p\u0026thinsp;\u0026lt;\u0026thinsp;0.05.\u003c/p\u003e\u003c/p\u003e\u003c/div\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eRadio-resistant A549 cell line showed cancer stem-like nature:\u0026nbsp;\u003c/strong\u003eTo identify gene signatures of radio-resistant lung cancer cells, we first generated the radio-resistant (RR) cells, which were evaluated for their radio-resistance by clonogenic assay after treatment with increasing doses (2-10 Gy) of radiation (Figure 2a,b).We found higher survival of RR cells after \u0026gamma;-irradiation, whereas the CC and R cells showed lower clonogenic survival. Since RR cells showed highest radio-resistance, further experiments were conducted using RR cells. Additionally, the radio-resistant nature of RR cells was also evaluated by the measurement of magnitude of radiation-induced DNA double strand break using \u0026gamma;H2AX. In response to \u0026gamma;-irradiation, % of \u0026gamma;H2AX positive cells was significantly lower in RR cells than CC cells at 3, 6 and 18 h of post-irradiation (Figure 2c) suggesting radio-resistant nature of RR cells. Presence of cancer stem cells is one of the well-known variables to determine the magnitude of radio-resistance \u003csup\u003e6, 7\u003c/sup\u003e. As RR cells were resistant to \u0026gamma;-irradiation, the expression of pluripotent genes (Oct4, Nanog, Sox2, TERT and CD166) were examined in CC and RR cells. Results showed significantly higher level of expression of these genes in RR cells than CC cells (Figure 3 a and b).Out of genes studied, Oct 4 and TERT showed highest fold change in expression in RR cells. Two of the pluripotent genes (Oct-4 and Nanog) were also evaluated using western blot which further corroborates the RT-PCR data (Supplementary Figure S3a). CSCs are known to form clone from single cell due to their self-renewal ability. Hence, to further validate the stemness feature in RR cells, the clone forming abilitywas evaluated employing single cell assay. We found that clone forming ability of RR cells was significantly higher than CC cells (Figure 3 c). CD44\u003csup\u003e+\u003c/sup\u003eCD24\u003csup\u003e-\u003c/sup\u003eis one of the hallmark features of CSCs, hence magnitude of CD44\u003csup\u003e+\u003c/sup\u003eCD24\u003csup\u003e-\u003c/sup\u003ewas quantified in RR and CC cells. Our results showed about 3 folds higher fraction of CD44\u003csup\u003e+\u003c/sup\u003eCD24\u003csup\u003e-\u003c/sup\u003ein cultures of RR cells (~75 %) than CC cells(~25 %) (Figure 3d,e). These set of resultssuggest higher CSC-like features of RR cells, which might be contributing to resistance nature of RR cells.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTranscriptomic profile of A549 RR cell lines\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNGS analysis was performed in RR cells and compared to the respective CC cells as mentioned in materials and methods. Compared to their respective controls, the number of differential expressed genes (DEGs) at FDR\u0026lt;0.05 were found to be 658 genes. Further analysis (log2 fold change \u0026gt; 1.3) of RR versus CC showed that 157 genes were up-regulated and 293 were down-regulated (Figure 4a). Gene enrichment analysis was performed using Panther Gene Ontology (GO) and the top 3 up/down-regulated processes with highest enrichment scoresare listed in Figure 4b. Our study showed enrichment of cell adhesion pathways in RR cells than controls suggesting higher stemness in RR cells. \u0026nbsp;Higher enrichment score of blood vessel development pathways in RR cells seems to be associated with increased angiogenic features due to CSCs. A plasticity of cancer cells towards cancer stem cells would result in de-differentiation and thus down-regulation of differentiation processes\u003csup\u003e25\u003c/sup\u003e. A549 cells being epithelial in nature, the down-regulation of fat cell and epithelial cell differentiation processes \u0026nbsp;(enrichment score 7.63 and 3.28, respectively) in RR cells is in line of their higher CSC like features of RR cells as observed in Figure 3 a-d. De-regulatedexpression of cell adhesion integrin subunits has been reported in breast and lung cancer spheroids enriched with CSCs and the adhesion in these cancer cells has been inhibited by antagonists of integrins\u003csup\u003e26\u003c/sup\u003e. It was also shown that up-regulation of COL17A1, a cell-adhesion molecule in colorectal cancer stem cells in organoid model, which become more sensitive to chemotherapy when COL17A1 was knocked out\u003csup\u003e27\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe volcano plots (Figure 4c) and heat maps (Figure 4d) of RR versus CC cells are also shown to compare and represent the expression profile of genes.The top 15 up- and down-regulated genes in RR versus CC cells were listed in Table 1. Out of the analysis, some of the genes which are up-regulated in cells obtained from RR cells were GRHL3, KRT19 and ALDH1A3. These genes were known to be expressed in cancer stem cells\u003csup\u003e28, 29, 30\u0026nbsp;\u003c/sup\u003esuggesting enrichment of CSC features in RR cells. \u0026nbsp;Further, NGS data in these cell lines were validated by real time PCR for the selected genes ALDH1A3, H19 and GHRL3 in samples of RR cell lines(Figure 3d).Genes like NID2, LGR5, and VCAN were found to be down-regulated. The DEGs among RR cells which were significantly up-regulated in this study were FHL1, LAMA4, CDCP1, ALX1, PCDH7, SAA1 and SUSD2. The genes, which were significantly down-regulated, were TRPA1, KDM5D, TTTY15, CLDN2, ANXA13, RHOBTB1, LIMCH1 and RBP4.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCorrelation of transcriptomic data of RR cells with radiotherapy treated lung cancer responders and non-responder patients selected from TCGA database\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo identify the signature genes from the transcriptomic data of RR versus CC groups, genes that are expressed in in-vitro system were correlated with patient data from TCGA database. Analysis showed that 60 genes were significantly correlated with selected patient\u0026rsquo;s transcriptomic data.To evaluate the above short-listed 60 individual genes to predict the RT outcome in lung cancer patients, binary logistic regression was applied. Out of these 60 genes, 7 genes (KCNB1, UNC13A, RIMS2, KCNH3, TOX2, SYTL3 and NR3C2) showed significant association (p\u0026lt;0.05) in predicting RT response in lung cancer patients (Table 2). One of the seven genes NR3C2 was validated using semi-quantitative RT-PCR to confirm transcriptomic data (Supplementary Figure S3b)\u003c/p\u003e\n\u003cp\u003eFurther, analysis was carried out using machine learning algorithms to evaluate the ability of these 7 genes together in predicting the RT outcome. Non-responders (n=29) and responders (n=21) data was divided in to training (80%) and testing (20%) sets followed by analysis using machine learning tools in Matlab 2019b.Application for all the available classification algorithms models was applied to the training data set for the best fit. KNN cosine model has shown higher accuracy (76%) with AUC of 0.80 (Figure 5a).Upon testing of the model, the accuracy was found to be 90% in predicting RT response. To evaluate the predictive ability of these gene signatures, nomogram analysis was performed based on ability of gene signature to predict responders and non-responders to RT. The results of KNN cosine model were used to generate the nomogram. The positive likelihood ratio of 7 gene signature was found to be 3.45 suggesting their predictive ability for RT response in lung cancer patients (Figure 5 b).\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eTo identify genes signature associated with stemness, which can predict RT response in lung adenocarcinoma patients, initially RR cell line was generated by treating clonogenic cells with a dose of 5 Gy/fraction eight times. The stem cell genes like Oct4, Nanog, Sox2, TERT and CD166 were highly up-regulated compared to the parental CC cells. Further these results were corroborated by significantly higher percent clone forming ability and expression of CD44\u003csup\u003e+\u003c/sup\u003eCD24\u003csup\u003e-\u003c/sup\u003e markers in RR cells suggesting stem like nature of these cells. After confirmation of RR cells for their stem-like nature, cells were evaluated for transcriptomic profiles of CSCs. Analysis of data has shown that up-regulated genes in this study like RIMS2, SFRP5, CCL5, ALDH1A3, H19 and VIPR1 and down-regulated genes like EIF1AY, RARB, HNF4A and SLC27A2 have been shown to be differentially expressed in CSCs of different tumours\u003csup\u003e31-36\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFurthermore, this study also focussed on transcriptomic data of RR versus CC cells and evaluated for genes which can predict RT response in lung adenocarcinoma patients. Total of 60 genes have been shortlisted, which has significant correlation between in-vitro samples and selected patient\u0026rsquo;s transcriptomic data. Among these, seven genes have been shortlisted which showed significant association in explaining the RT response by binary logistic regression. Out of these 7 genes, 5 genes (KCNB1, UNC13A, RIMS2, KCNH3 and TOX2) were up-regulated and 2 genes (SYTL3 and NR3C2) were down-regulated.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eKCNB1 expression was shown to be up-regulated in pancreatic cancer stem cells and blocking this potassium voltage-gated channel with 4-aminopyridine inhibited the proliferation of the cells. This was also proposed to be a potential target for pancreatic cancer \u003csup\u003e37\u003c/sup\u003e. In our study, we have shown that KCNB1 gene is highly up regulated in radio-resistant cell line. Additionally, our study has shown its usefulness as a signature gene to predict RT response in lung adenocarcinoma patients. \u0026nbsp;It was reported earlier that UNC13A was shown to be having single nucleotide polymorphisms in keratinocyte carcinomas and this gene\u0026rsquo;s role in reprogramming human astrocytes into functional neurons through notch signalling\u003csup\u003e38, 39\u003c/sup\u003e. In our study, UNC13A gene expression is up-regulated in radio-resistant cell line and patients\u0026rsquo; samples.\u003c/p\u003e\n\u003cp\u003eThe genes RIMS2 and KCNH3 expression was shown to be up-regulated in various other cancers\u003csup\u003e40-43\u003c/sup\u003e. In contrast some groups have shown the down-regulation of RIMS2 in glioma and breast cancers\u003csup\u003e44, 45\u003c/sup\u003e. Additionally, it was also shown RIMS2 is highly up-regulated in mesenchymal stem cells of umbilical cord blood\u003csup\u003e46\u003c/sup\u003e. Our data showed that RIMS2 was up-regulated in transcriptomic data of RR cells and patient samples. Its role in radio-resistance is yet to be evaluated. KCNH3 was shown to be highly expressed in ovarian cancers and cardiac progenitor cells. It was suggested that high expression of this gene might be an independent predictor of poor overall disease free survival\u003csup\u003e47-49\u003c/sup\u003e. Our data showed that KCNH3 gene is highly up regulated in radio-resistant cell line and this gene has potential in differentiating RT responders from non-responders.\u003c/p\u003e\n\u003cp\u003eThe gene TOX2 expression was earlier shown to be associated with worst overall survival of natural killer/T cell lymphoma patients\u003csup\u003e50\u003c/sup\u003e. It is also showed that its role in RUNX3-TOX2-Super Enhancer-TOX2-Protein of regenerating liver 3 regulatory pathway of natural killer T cell lymphoma biology.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eHowever, its role in radio-resistance and stemness is not clearly established. Here in our study, TOX2 is highly expressed in radio-resistant cell lines of A549 and has ability in differentiating RT response of lung adenocarcinoma patients (Table 2). Furthermore, it would be very interesting to evaluate these genes (KCNB1, UNC13A, RIMS2, KCNH3 and TOX2) role in radiation resistant tumors.\u003c/p\u003e\n\u003cp\u003eNR3C2 and SYTL3 were shown to be down-regulated in different cancers\u003csup\u003e51, 52\u003c/sup\u003e. Our data also showed that NR3C2 and SYTL3 genes were down-regulated in RR versus CC cells and patient samples. SYTL3 expression and its functional evaluation have not been done in cancers but its knockdown is correlated with the neuronal cell migration in embryonic mouse cortex. When all these seven genes were used together to predict the RT response using machine learning algorithms, it showed potential in differentiating RT non-responders from responders.\u003c/p\u003e\n\u003cp\u003eIn conclusion, this study brings out valuable information about the genes relevant to CSC like cells and possible role in resistance nature of lung carcinoma cells. Additionally seven genes signature was obtained from correlating in-vitro RR cell line model and patient\u0026rsquo;s data. In this effort, the predictive ability of the signature genes helps in identifying non-responders and before translating them to clinics it is needed to be evaluated in a separate prospective study in radio-therapy treated lung cancer patients. Finally, the functional evaluation of these genes in the proposed signature needs to be evaluated for their ability to sensitize the RR lung cancer cell models which may have applicability in future RT treatment protocols.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eFunding Declaration:\u0026nbsp;\u003c/strong\u003eAuthors acknowledge Bhabha Atomic Research Centre, Department of Atomic Energy, Government of India for financial support of this work.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eM.M.S.B conducted the experiments, analyzed the results, designed the study, wrote and edited the manuscript, P.K.M helped in analysis of data, writing and editing the manuscript, D.D has analyzed data using machine learning tools in Python, S.V.R and VG conducted part of the experiments in the manuscript, H.S.B, S.M, N.G, N.V.S have helped in conducting the RNA Seq protocols on the samples. B.N. P analysed the results, designed the study and edited the manuscript.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eRaw data files have been submitted GEO website database (https://www.ncbi.nlm.nih.gov/geo) and the accession number is GSE221796\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eSung, H. et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. : 71: 209\u0026thinsp;\u0026ndash;\u0026thinsp;49.doi: 10.3322/caac.21660. Epub 2021 Feb 4. PMID: 33538338. (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTerada, Y. et al. Radiotherapy for local recurrence of non-small-cell lung cancer after lobectomy and lymph node dissection-can local recurrence be radically cured by radiation? Jpn J Clin Oncol. ;50(4):425\u0026thinsp;\u0026ndash;\u0026thinsp;33. (2020). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/jjco/hyz188\u003c/span\u003e\u003cspan address=\"10.1093/jjco/hyz188\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 31926488.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHerbst, R. S., Morgensztern, D. \u0026amp; Boshoff, C. The biology and management of non-small cell lung cancer. Nature. ; 553(7689):446\u0026thinsp;\u0026ndash;\u0026thinsp;54. (2018). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/nature25183\u003c/span\u003e\u003cspan address=\"10.1038/nature25183\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 29364287.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLe Chevalier, T. et al. Radiotherapy alone versus combined chemotherapy and radiotherapy in nonresectable non-small-cell lung cancer: First analysis of a randomized trial in 353 patients. J. Natl. Cancer Inst. 83, 417\u0026ndash;23. (1991). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/jnci/83.6.417\u003c/span\u003e\u003cspan address=\"10.1093/jnci/83.6.417\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 1847977.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCurran, W. J. Jr. et al. Sequential vs. concurrent chemoradiation for stage III non-small cell lung cancer: Randomized phase III trial RTOG 9410. \u003cem\u003eJ. Natl. Cancer Inst.\u003c/em\u003e \u003cb\u003e103\u003c/b\u003e, 1452\u0026ndash;1460. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/jnci/djr325\u003c/span\u003e\u003cspan address=\"10.1093/jnci/djr325\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2011).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNguyen, L. V., Vanner, R., Dirks, P. \u0026amp; Eaves, C. J. Cancer stem cells: an evolving concept. Nat Rev Cancer. ; 12(2):133\u0026thinsp;\u0026ndash;\u0026thinsp;43. (2012). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/nrc3184\u003c/span\u003e\u003cspan address=\"10.1038/nrc3184\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 22237392.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBaccelli, I. \u0026amp; Trumpp, A. The evolving concept of cancer and metastasis stem cells. \u003cem\u003eJ. Cell. Biol.\u003c/em\u003e \u003cb\u003e198\u003c/b\u003e (3), 281\u0026ndash;293. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1083/jcb.201202014\u003c/span\u003e\u003cspan address=\"10.1083/jcb.201202014\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2012). PMID: 22869594; PMCID: PMC3413352.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSant, G. R., Knopf, K. B. \u0026amp; Albala, D. M. Live-single-cell phenotypic cancer biomarkers-future role in precision oncology? NPJ Precis Oncol. ;1(1):21. (2017). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41698-017-0025-y\u003c/span\u003e\u003cspan address=\"10.1038/s41698-017-0025-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 29872705.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eStastna, M. \u0026amp; Van Eyk, J. E. Secreted proteins as a fundamental source for biomarker discovery. Proteomics. ;12(4\u0026ndash;5):722\u0026thinsp;\u0026ndash;\u0026thinsp;35. doi: 10.1002/pmic.201100346. Epub 2012 Jan 19. PMID: 22247067 (2012).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBeer, D. G. et al. Gene-expression profiles predict survival of patients with lung adenocarcinoma. \u003cem\u003eNat. Med.\u003c/em\u003e \u003cb\u003e8\u003c/b\u003e, 816\u0026ndash;824. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/nm733\u003c/span\u003e\u003cspan address=\"10.1038/nm733\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2002). Epub 2002 Jul 15. PMID: 12118244.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eShedden, K. et al. Gene expression-based survival prediction in lung adenocarcinoma: A multi-site, blinded validation study. \u003cem\u003eNat. Med.\u003c/em\u003e \u003cb\u003e14\u003c/b\u003e, 822\u0026ndash;827. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/nm.1790\u003c/span\u003e\u003cspan address=\"10.1038/nm.1790\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2008). Epub 2008 Jul 20. PMID: 18641660; PMCID: PMC2667337.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi, Y. et al. Network-based approach identified cell cycle genes as predictor of overall survival in lung adenocarcinoma patients. \u003cem\u003eLung Cancer\u003c/em\u003e. \u003cb\u003e80\u003c/b\u003e, 91\u0026ndash;98 (2013). Epub 2013 Jan 26. PMID: 23357462; PMCID: PMC3595338.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBianchi, F. et al. Survival prediction of stage I lung adenocarcinomas by expression of 10 genes. \u003cem\u003eJ. Clin. Investig\u003c/em\u003e. \u003cb\u003e117\u003c/b\u003e, 3436\u0026ndash;3444 (2007). PMID: 17948124; PMCID: PMC2030461.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRaz, D. J. et al. A multigene assay is prognostic of survival in patients with early-stage lung adenocarcinoma. Clin. Cancer Res. 14, 5565\u0026ndash;70. (2008). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1158/1078-0432.CCR-08-0544\u003c/span\u003e\u003cspan address=\"10.1158/1078-0432.CCR-08-0544\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 18765549.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePark, Y. Y. et al. Development and validation of a prognostic gene-expression signature for lung adenocarcinoma. PLoS ONE 7, e44225. (2012). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1371/journal.pone.0044225\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0044225\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Epub 2012 Sep 7. PMID: 22970185.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAkagi, I. et al. Combination of protein coding and noncoding gene expression as a robust prognostic classifier in stage I lung adenocarcinoma. \u003cem\u003eCancer Res.\u003c/em\u003e \u003cb\u003e73\u003c/b\u003e, 3821\u0026ndash;3832. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1158/0008-5472.CAN-13-0031\u003c/span\u003e\u003cspan address=\"10.1158/0008-5472.CAN-13-0031\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2013). Epub 2013 May 2. PMID: 23639940.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTomida, S. et al. Relapse-related molecular signature in lung adenocarcinomas identifies patients with dismal prognosis. J. Clin. Oncol. 27, 2793\u0026ndash;99.doi: 10.1200/JCO.2008.19.7053. Epub 2009 May 4. PMID: 19414676. (2009).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLarsen, J. E. et al. Gene expression signature predicts recurrence in lung adenocarcinoma. Clin. Cancer Res. 13, 2946\u0026ndash;54. (2007). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1158/1078-0432.CCR-06-2525\u003c/span\u003e\u003cspan address=\"10.1158/1078-0432.CCR-06-2525\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 17504995.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePeinado-Serrano, J. et al. A Six-Gene Prognostic and Predictive Radiotherapy-Based Signature for Early and Locally Advanced Stages in Non-Small-Cell Lung Cancer. \u003cem\u003eCancers (Basel)\u003c/em\u003e. \u003cb\u003e14\u003c/b\u003e (9), 2054. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/cancers14092054\u003c/span\u003e\u003cspan address=\"10.3390/cancers14092054\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2022). PMID: 35565183; PMCID: PMC9099638.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEke, I. et al. Radiotherapy alters expression of molecular targets in prostate cancer in a fractionation- and time-dependent manner. Sci Rep. ;12(1):3500. (2022). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41598-022-07394-y\u003c/span\u003e\u003cspan address=\"10.1038/s41598-022-07394-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 35241721.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBalla, M. M. S., Yadav, H. D. \u0026amp; Pandey, B. N. Tumorsphere assay provides a better in vitro method for cancer stem-like cells enrichment in A549 lung adenocarcinoma cells. \u003cem\u003eTissue Cell.\u003c/em\u003e \u003cb\u003e60\u003c/b\u003e, 21\u0026ndash;24 (2019). Epub 2019 Jul 16. PMID: 31582014.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBalla, M. M., Ningthoujam, R. S., Kumar, M., Bandekar, J. R. \u0026amp; Pandey, B. N. Cellular and spectroscopic characterization of cancer stem cell-like cells derived from A549 lung carcinoma. J Cancer Res Ther. (2016). Jul-Sep;12(3):1144-52 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.4103/0973-1482.171365\u003c/span\u003e\u003cspan address=\"10.4103/0973-1482.171365\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 28054526.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSanchez-Vega, F. et al. Oncogenic Signaling Pathways in The Cancer Genome Atlas. \u003cem\u003eCell\u003c/em\u003e \u003cb\u003e173\u003c/b\u003e (2), 321\u0026ndash;337e10. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.cell.2018.03.035\u003c/span\u003e\u003cspan address=\"10.1016/j.cell.2018.03.035\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2018). PMID: 29625050; PMCID: PMC6070353.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBatut, B., van den Beek, M., Doyle, M. A. \u0026amp; Soranzo, N. RNA-Seq Data Analysis in Galaxy. Methods Mol Biol. ; 2284:367\u0026ndash;392. (2021). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/978-1-0716-1307-8_20\u003c/span\u003e\u003cspan address=\"10.1007/978-1-0716-1307-8_20\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 33835453.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVlashi, E. \u0026amp; Pajonk, F. Cancer stem cells, cancer cell plasticity and radiation therapy. Semin Cancer Biol. ;31:28\u0026ndash;35. (2015). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.semcancer.2014.07.001\u003c/span\u003e\u003cspan address=\"10.1016/j.semcancer.2014.07.001\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Epub 2014 Jul 12. PMID: 25025713.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePaolillo, M. et al. Stem-Like Cancer Cells in a Dynamic 3D Culture System: A Model to Study Metastatic Cell Adhesion and Anti-Cancer Drugs. Cells. ;8(11):1434. (2019). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/cells8111434\u003c/span\u003e\u003cspan address=\"10.3390/cells8111434\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 31766310.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eOhta, Y. et al. Cell-matrix interface regulates dormancy in human colon cancer stem cells. \u003cem\u003eNature\u003c/em\u003e \u003cb\u003e608\u003c/b\u003e (7924), 784\u0026ndash;794. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41586-022-05043-y\u003c/span\u003e\u003cspan address=\"10.1038/s41586-022-05043-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2022). Epub 2022 Jul 7. PMID: 35798028.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLin, Z. et al. Murine interfollicular epidermal differentiation is gradualistic with GRHL3 controlling progression from stem to transition cell states. Nat Commun. ;11(1):5434. (2020). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41467-020-19234-6\u003c/span\u003e\u003cspan address=\"10.1038/s41467-020-19234-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 33116143.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSullivan, K. E., Rojas, K., Cerione, R. A., Nakano, I. \u0026amp; Wilson, K. F. The stem cell/cancer stem cell marker ALDH1A3 regulates the expression of the survival factor tissue transglutaminase, in mesenchymal glioma stem cells. \u003cem\u003eOncotarget\u003c/em\u003e \u003cb\u003e8\u003c/b\u003e (14), 22325\u0026ndash;22343. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.18632/oncotarget.16479\u003c/span\u003e\u003cspan address=\"10.18632/oncotarget.16479\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2017). PMID: 28423611; PMCID: PMC5410226.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLecerf, C., Peperstraete, E., Le Bourhis, X. \u0026amp; Adriaenssens, E. Propagation and Maintenance of Cancer Stem Cells: A Major Influence of the Long Non-Coding RNA H19. \u003cem\u003eCells\u003c/em\u003e \u003cb\u003e9\u003c/b\u003e (12), 2613. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/cells9122613\u003c/span\u003e\u003cspan address=\"10.3390/cells9122613\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2020). PMID: 33291403; PMCID: PMC7762009.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSecco, M. et al. Gene expression profile of mesenchymal stem cells from paired umbilical cord units: cord is different from blood. Stem Cell Rev Rep. ;5(4):387\u0026ndash;401. (2009). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s12015-009-9098-5\u003c/span\u003e\u003cspan address=\"10.1007/s12015-009-9098-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 20058202.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBehera, A., Ashraf, R., Srivastava, A. K. \u0026amp; Kumar, S. Bioinformatics analysis and verification of molecular targets in ovarian cancer stem-like cells. \u003cem\u003eHeliyon\u003c/em\u003e \u003cb\u003e6\u003c/b\u003e (9), e04820. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.heliyon.2020.e04820\u003c/span\u003e\u003cspan address=\"10.1016/j.heliyon.2020.e04820\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2020). PMID: 32984578; PMCID: PMC7492822.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKawakami, K. et al. Secreted frizzled-related protein-5 is epigenetically downregulated and functions as a tumor suppressor in kidney cancer. \u003cem\u003eInt. J. Cancer\u003c/em\u003e. \u003cb\u003e128\u003c/b\u003e (3), 541\u0026ndash;550. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/ijc.25357\u003c/span\u003e\u003cspan address=\"10.1002/ijc.25357\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2011). PMID: 20340127; PMCID: PMC3940353.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHuang, R. et al. CCL5 derived from tumor-associated macrophages promotes prostate cancer stem cells and metastasis via activating β-catenin/STAT3 signaling. \u003cem\u003eCell. Death Dis.\u003c/em\u003e \u003cb\u003e11\u003c/b\u003e (4), 234doi. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41419-020-2435-y\u003c/span\u003e\u003cspan address=\"10.1038/s41419-020-2435-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2020). PMID: 32300100; PMCID: PMC7162982.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSastry, K. S. et al. Cytoprotective effect of neuropeptides on cancer stem cells: vasoactive intestinal peptide-induced antiapoptotic signaling. \u003cem\u003eCell. Death Dis.\u003c/em\u003e \u003cb\u003e8\u003c/b\u003e (6), e2844. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/cddis.2017.226\u003c/span\u003e\u003cspan address=\"10.1038/cddis.2017.226\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2017). PMID: 28569785; PMCID: PMC5520887.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSu, J. et al. SLC27A2 induces cisplatin resistance in lung cancer stem cells by negatively regulating Bmi1-ABCG2 signaling. \u003cem\u003eMol. Carcinog.\u003c/em\u003e \u003cb\u003e55\u003c/b\u003e (11), 1822\u0026ndash;1832. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/mc.22430\u003c/span\u003e\u003cspan address=\"10.1002/mc.22430\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2016). Epub 2015 Oct 29. PMID: 26513225.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eShiozaki, A. et al. Roles of voltage\u0026ndash;gated potassium channels in the maintenance of pancreatic cancer stem cells. \u003cem\u003eInt. J. Oncol.\u003c/em\u003e \u003cb\u003e59\u003c/b\u003e (4), 76. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3892/ijo.2021.5256\u003c/span\u003e\u003cspan address=\"10.3892/ijo.2021.5256\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2021). Epub 2021 Aug 20. PMID: 34414448.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSunder-Plassmann, R., Geusau, A., Endler, G., Weninger, W. \u0026amp; Wielscher, M. Identification of Genetic Risk Factors for Keratinocyte Cancer in Immunosuppressed Solid Organ Transplant Recipients: A Case-Control Study. Cancers (Basel). ;15(13):3354. (2023). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/cancers15133354\u003c/span\u003e\u003cspan address=\"10.3390/cancers15133354\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 37444464.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYin, J. C. et al. Chemical Conversion of Human Fetal Astrocytes into Neurons through Modulation of Multiple Signaling Pathways. Stem Cell Reports. ;12(3):488\u0026ndash;501. doi: 10.1016/j.stemcr.2019.01.003. Epub 2019 Feb 7. PMID: 30745031 (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSecco, M. et al. Gene expression profile of mesenchymal stem cells from paired umbilical cord units: cord is different from blood. Stem Cell Rev Rep. ;5(4):387\u0026ndash;401. (2009). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s12015-009-9098-5\u003c/span\u003e\u003cspan address=\"10.1007/s12015-009-9098-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 20058202.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBehera, A., Ashraf, R., Srivastava, A. K. \u0026amp; Kumar, S. Bioinformatics analysis and verification of molecular targets in ovarian cancer stem-like cells. \u003cem\u003eHeliyon\u003c/em\u003e \u003cb\u003e6\u003c/b\u003e (9), e04820. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.heliyon.2020.e04820\u003c/span\u003e\u003cspan address=\"10.1016/j.heliyon.2020.e04820\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2020). PMID: 32984578; PMCID: PMC7492822.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVezzali, R. et al. The FOXG1/FOXO/SMAD network balances proliferation and differentiation of cortical progenitors and activates Kcnh3 expression in mature neurons. Oncotarget. ;7(25):37436\u0026ndash;37455. (2016). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.18632/oncotarget.9545\u003c/span\u003e\u003cspan address=\"10.18632/oncotarget.9545\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 27224923.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi, Z. et al. KCNH3 Predicts Poor Prognosis and Promotes Progression in Ovarian Cancer. Onco Targets Ther. ;13:10323\u0026ndash;10333. (2020). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.2147/OTT.S268055\u003c/span\u003e\u003cspan address=\"10.2147/OTT.S268055\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 33116612.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWei, B., Wang, L. \u0026amp; Zhao, J. Circular RNA hsa_circ_0005114-miR-142-3p/miR-590-5p-adenomatous polyposis coli protein axis as a potential target for treatment of glioma. \u003cem\u003eOncol. Lett.\u003c/em\u003e \u003cb\u003e21\u003c/b\u003e (1), 58doi. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3892/ol.2020.12320\u003c/span\u003e\u003cspan address=\"10.3892/ol.2020.12320\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2021). Epub 2020 Nov 19. PMID: 33281969; PMCID: PMC7709550.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhang, L., Liu, Z. \u0026amp; Zhu, J. In silico screening using bulk and single-cell RNA-seq data identifies RIMS2 as a prognostic marker in basal-like breast cancer: A retrospective study. \u003cem\u003eMed. (Baltim).\u003c/em\u003e \u003cb\u003e100\u003c/b\u003e (16), e25414 (2021). PMID: 33879671; PMCID: PMC8078249.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSecco, M. et al. Gene expression profile of mesenchymal stem cells from paired umbilical cord units: cord is different from blood. Stem Cell Rev Rep. ;5(4):387\u0026ndash;401. (2009). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s12015-009-9098-5\u003c/span\u003e\u003cspan address=\"10.1007/s12015-009-9098-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 20058202.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhang, Y. et al. Epigenomic Reprogramming of Adult Cardiomyocyte-Derived Cardiac Progenitor Cells. Sci Rep. ;5:17686. (2015). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/srep17686\u003c/span\u003e\u003cspan address=\"10.1038/srep17686\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Erratum in: Sci Rep. 2017;7:46907. PMID: 26657817.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVezzali, R. et al. The FOXG1/FOXO/SMAD network balances proliferation and differentiation of cortical progenitors and activates Kcnh3 expression in mature neurons. Oncotarget. ;7(25):37436\u0026ndash;37455. (2016). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.18632/oncotarget.9545\u003c/span\u003e\u003cspan address=\"10.18632/oncotarget.9545\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 27224923.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi, Z. et al. KCNH3 Predicts Poor Prognosis and Promotes Progression in Ovarian Cancer. Onco Targets Ther. ;13:10323\u0026ndash;10333. (2020). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.2147/OTT.S268055\u003c/span\u003e\u003cspan address=\"10.2147/OTT.S268055\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 33116612.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhou, J. et al. Super-enhancer-driven TOX2 mediates oncogenesis in Natural Killer/T Cell Lymphoma. Mol Cancer. ;22(1):69. (2023). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s12943-023-01767-1\u003c/span\u003e\u003cspan address=\"10.1186/s12943-023-01767-1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 37032358.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi, X. et al. Nuclear receptor subfamily 3 group c member 2 (NR3C2) is downregulated due to hypermethylation and plays a tumor-suppressive role in colon cancer. \u003cem\u003eMol. Cell. Biochem.\u003c/em\u003e \u003cb\u003e477\u003c/b\u003e (11), 2669\u0026ndash;2679. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s11010-022-04449-6\u003c/span\u003e\u003cspan address=\"10.1007/s11010-022-04449-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2022). Epub 2022 May 23. PMID: 35604518.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDong, X. et al. Transcriptional networks identify synaptotagmin-like 3 as a regulator of cortical neuronal migration during early neurodevelopment. Cell Rep. ; 34(9):108802. (2021). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.celrep.2021.108802\u003c/span\u003e\u003cspan address=\"10.1016/j.celrep.2021.108802\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 33657377.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003eTable 1 and 2 are available in the Supplementary Files section.\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-7240316/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7240316/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eDespite major technical advancements, the prediction of radiotherapy clinical outcome is a major challenge to radiation oncologists due to lack of suitable predictive biomarkers. To address this, a radio-resistant cell line (RR) has been generated from A549 cells (CC) after fractionated doses followed by clonogenic assay. On evaluation these cells showed cancer stem like features. The transcriptomic data of RR versus CC cells were analysed. Out of 658 differentially expressed genes, RNA expression of 60 genes were found to be correlated (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05) with transcriptomic data of radiotherapy treated lung adenocarcinoma patients obtained from the TCGA database. Binary logistic regression of these 60 selected genes resulted in identification of seven genes (KCNB1, UNC13A, RIMS2, KCNH3, TOX2, SYTL3 and NR3C2) which showed significant (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05) association with response to radiotherapy. Data was employed to predict radiotherapy response in patients using machine learning algorithms [KNN Cosine]. Algorithms were trained on 80% and tested on 20% of patient\u0026rsquo;s data. Accuracy was 90% for the model in predicting radiotherapy outcome. When nomogram analysis was performed based on the results of KNN Cosine model, it showed a positive likelihood ratio of 3.45, suggesting potential prognostic nature of this gene signature for radiotherapy outcome in lung cancer patients.\u003c/p\u003e","manuscriptTitle":"Transcriptomic Analysis of Radio-resistant A549 Cells Identify Stemness Gene Signature to Predict Radiotherapy Response in Lung Adenocarcinoma Patients","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-09 16:05:19","doi":"10.21203/rs.3.rs-7240316/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"db98e166-2726-4183-992d-bea85713908e","owner":[],"postedDate":"September 9th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":54309799,"name":"Health sciences/Biomarkers"},{"id":54309800,"name":"Biological sciences/Cancer"},{"id":54309801,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":54309802,"name":"Biological sciences/Genetics"},{"id":54309803,"name":"Health sciences/Oncology"}],"tags":[],"updatedAt":"2025-09-26T18:23:12+00:00","versionOfRecord":[],"versionCreatedAt":"2025-09-09 16:05:19","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7240316","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7240316","identity":"rs-7240316","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}