Research ArticleExperimental Studies
Open Access

Transcriptome Analysis of the Effects of X-Ray Radiotherapy on Non-small-cell Lung Cancer Using Next-generation Sequencing

SERHAT ARAS, TUĞBA KUL KÖPRÜLÜ, BURÇIN ERKAL ÇAM, JÜLIDE BALKAN, ESRA ERDEM, ESRA ÇIKLER and ALTAN KARA
In Vivo September 2025, 39 (5) 2711-2727; DOI: https://doi.org/10.21873/invivo.14070

Abstract

Background/Aim: This study investigated the acute effects of flattening filter (FF) and flattening filter-free (FFF) beams on gene expression in non-small-cell lung cancer (NSCLC).

Materials and Methods: Thirty-six adult athymic nude mice were divided into five groups. The control group did not undergo any radiotherapy or treatment procedures, whereas in the lung cancer (LCa) group, a cancer model was created but not irradiated. LCa models received 20 Gy radiotherapy with FF at 400 MU/min, or with FFF at 1,000 or 1,800 MU/min dose rates. The mice were irradiated 20 days after A549 cancer cell-line implantation and sacrificed 48 h after irradiation for genetic analysis.

Results: Twelve genes were identified as being common across all radiotherapy groups. The expression of most of these genes changed as the dose rate increased. Seven of these genes were also common to the LCa and control groups. Three genes down-regulated in the untreated cancer group showed increased expression with higher dose rates in treated groups. Significant differences were observed in glutamatergic synapse, actin cytoskeleton regulation, and steroid synthesis in FF-400 and FFF-1000. The FFF-1800 group exhibited significant changes in RNA transport, actin cytoskeleton regulation, and phagosome-associated pathways.

Conclusion: FFF beams induced more extensive and pronounced gene-expression changes compared to FF beams in NSCLC.

Keywords:

Introduction

The prevention and treatment of lung cancer (LCa) is of great importance due to its high incidence and mortality (1). Modern therapy techniques are commonly used to achieve this goal. Despite significant advancements in the treatment of LCa through surgery, radiotherapy, and chemotherapy, the long-term prognosis for patients remains unsatisfactory due to various factors (2). During cancer treatment, such as for lung, head and neck, breast, and prostate cancer, radiotherapy aims to deliver a precise and controlled lethal dose to tumor tissue while minimizing exposure of healthy tissues outside the targeted area (3, 4). Although LCa treatment is a highly complex and multidisciplinary approach, significant progress has been achieved in the past decade with the development of new technology and treatment techniques (5).

One such development increasingly applied in cancer radiotherapy (6) is that of flattening filter-free (FFF) high instantaneous dose-rate beams, which allow application of up to a fourfold increase in dose rate compared to conventional flattening filter (FF) low-dose rate beams. Thanks to FFF beams, high dose rates are frequently used in clinical applications of radiotherapy to reduce treatment times (7). Furthermore, high-dose rate beams are becoming increasingly popular in stereotactic body radiation therapy, stereotactic radiosurgery, and volume-modulated arc therapy techniques due to their ability to significantly reduce patient irradiation times (8, 9).

Although the curative effect of radiotherapy in cancer treatments has been known for a long time, a better understanding of how tumors respond to radiation is still required. Next-generation sequencing (NGS) is a powerful technology that requires bioinformatic analysis for data interpretation and has increased in importance in recent years in the field of genomics by enabling the sequencing of large amounts of DNA or RNA (10, 11). NGS can be used to identify specific genetic alterations that may affect tumor response to radiotherapy, in prediction of tumor radiosensitivity. It provides information on gene-expression levels and other genomic features that may be associated with radiosensitivity. Despite the widespread use of NGS in oncology, its clinical applications in radiation oncology are still limited. NGS is not yet widely used in the context of predicting tumor radiosensitivity or guiding radiation therapy decisions (12).

Non-small-cell lung cancer (NSCLC) is commonly treated using high-dose rate beams (13). However, there are insufficient studies in the literature on the radiobiological effects of these FFF high instantaneous dose-rate beams on the acute NGS-based radiotherapy-response induced in NSCLC.

In the present study, NSCLC models were established in outbred athymic nude mice and single-dose 20 Gy radiotherapy was applied at different dose rates. The study analyzed changes in gene-expression levels induced by radiotherapy using NGS technology. The aim of this study was to analyze the effect of changes in tumoral radiotherapy-response to investigate the acute radiobiological mechanisms underlying LCa inhibition after irradiation using FF and FFF beams in nude mouse models.

Materials and Methods

Experimental animals and treatment groups. Adult female outbred athymic nude mice (strain #:007850) weighing 20-30 g and aged 8 weeks were obtained from Yeditepe University Animal Experiments Laboratory. The mice were housed in standard laboratory conditions with a constant temperature of 23±2°C, a humidity of 60±5%, and a ventilated and sunlit environment. The mice were kept in pathogen-free conditions in a laminar flow cabinet according to established and approved protocols. They were housed in HEPA-filtered rooms and IVC cages with their own ventilation systems. Sterile sawdust was used as litter. The mice were provided with sterile feed ad libitum and housed in a room with a 12-hour light and 12-hour dark cycle. The humidity, ventilation, and temperature of the rooms were controlled daily using an automation system. The cage and litter equipment were autoclaved. Ethics Committee approval was given by Yeditepe University Faculty of Medicine Experimental Research Center for the care and use of experimental animals (approval number: 2023/01-04).

The mice were randomly divided into five experimental groups:

  • A1 (Control): No action was applied to this group of mice (n=4).

  • A2 (LCa): LCa models were created in this group of mice (n=8).

  • A3 (FF-400): LCa models were created and radiotherapy was applied to a single dose of 20 Gy using a dose rate of 400 MU/min (n=8).

  • A4 (FFF-1000): LCa models were created and radiotherapy was applied to a single dose of 20 Gy at a dose rate of 1000 MU/min (n=8).

  • A5 (FFF-1800): LCa models were created and radiotherapy was applied to a single dose of 20 Gy at a dose rate of 1800 MU/min (n=8).

Cell culture and in-vivo xenograft mouse model of NSCLC. A549 NSCLC cells were obtained from the American Type Culture Collection (CRL-6475™; American Type Culture Collection, Manassas, VA, USA) for cell culture procedures. The cells were incubated in High Glucose Dulbecco’s Modified Eagle’s Medium (DMEM-HG) (Gibco™, Waltham, MA, USA) medium containing a final concentration of 2% penicillin/streptomycin (Sigma-Aldrich, Darmstadt, Germany) and 10% fetal bovine serum (Sigma-Aldrich) at 37°C with 5% CO2 for 24 h. When the cell culture dish reached 80% confluency, cells were passaged using trypsin (Sigma-Aldrich). A mixture of 1×107 cells and DMEM-HG:matrigel (1:1 ratio Corning® Matrigel® Matrix, Corning Inc., Corning, NY, USA) was injected subcutaneously into the right dorsal flank at a rate of 100 μl per animal. No swelling or fluid leakage was observed at the injection site. Tumor progression was evaluated on specific days following the injection, and tumor growth was monitored.

Radiotherapy application procedures. Radiotherapy was administered 20 days after tumor cell implantation, at which point tumors were visibly palpable and had reached approximately 5-8 mm in diameter, indicating established tumor growth. Except for the subjects in the normal control (A1) and LCa control (A2) groups, radiotherapy was performed under general anesthesia by administering 60 mg/kg ketamine and 8 mg/kg xylazine intraperitoneally to immobilize the mice during irradiation. The mice were then placed in a supine position on a plexiglass tray for radiotherapy. Each mouse was administered a single dose of 20 Gy total (equivalent to approximately 60-70 Gy in conventional human biological terms) using a 6-MV Varian Trilogy (Varian Medical System, Palo Alto, CA, USA) ionizing X-ray linear accelerator device. The dose rate was 400 MU/min in FF mode and 1,000 and 1,800 MU/min in FFF modes. The skin-source distance was 100 cm using the source-to-axis distance technique. A 10 mm bolus was placed on the radiotherapy field prior to radiotherapy to compensate for dose depth and distribution.

Euthanasia. Forty-eight hours after radiotherapy, nude mice were placed under general anesthesia with 60 mg/kg ketamine and 8 mg/kg intraperitoneally. The mice were euthanized by CO2 inhalation. LCa tissues (10-15 mg) were then surgically removed and preserved under appropriate conditions for the study of gene expression levels.

Total RNA isolation and determination of RNA yield. Total RNA was isolated from fresh tumor tissue samples in DNA/RNA Shield using RNeasy Mini Kit (Qiagen GmbH, Hilden, Germany) according to the manufacturer’s protocol. The tissue samples were first centrifuged, and the DNA/RNA shield was completely removed from the tubes containing approximately 10 mg of the tissue sample. RLT buffer (400 μl) was added to each sample, and tissue samples were disrupted and homogenized for 10 min at an oscillation of 50/s with TissueLyzer LT (Qiagen GmbH) Following homogenization, the supernatant was removed and placed into a new centrifuge tube. Once 70% ethanol was added to each tissue lysate, 700 μl of the mixture was loaded onto an RNeasy Mini spin column (Qiagen GmbH) and centrifuged for 15 s at 10,000 g. After washing processes, RNA was eluted from the silica membrane using 50 μL of RNAse-free water and stored at −20°C. The concentration and integrity of the RNA samples were determined using Qubit RNA Broad Range Assay Kit (Invitrogen, Carlsbad, CA, USA) with a Qubit 4.0 fluorometer and RNA ScreenTape Assay with TapeStation 4150 (Agilent Technologies, Santa Clara, CA, USA), respectively.

Library preparation and RNA sequencing. A total of 500 ng of each RNA sample was used for library preparation using Illumina Stranded Total RNA Prep, Ligation with Ribo-Zero Plus kit (Illumina, Inc., San Diego, CA, USA) following manufacturer recommendations. Briefly, ribosomal RNA (rRNA) was first depleted from the total RNA sample, and the rRNA-depleted RNA was fragmented. cDNA was synthesized by annealing random hexamers to a fragmented RNA sample. After creating double-stranded cDNA, unique barcoded adapters were attached to the RNA samples, and the cDNA fragments were then amplified via polymerase chain reaction. The concentration of each of the constructed libraries was measured using a Qubit 4.0 fluorometer with a Qubit RNA broad-range kit. Constructed libraries were combined in equimolar quantities. Sequencing of 2×150 bp paired-end reads was carried out using an S1 flowcell on Illumina NovaSeq 6000 Sequencing System (Illumina, Inc.).

Sequencing data analysis. Collected experimental data was converted to fastq format using bcl2fastq2, and the quality check was performed via FastQC. Sequence reads were aligned to C57BL_6NJ_v1(GCA_001632555.1) genome assembly by using HISAT2 and aligned files stored in BAM format. Required steps such as sorting and indexing the alignment files were performed by using samtools. Subsequently, count matrices for samples were generated by using htseq-count. At this stage, both the reference genome assembly and required GFF file for annotation were downloaded from Ensembl database and count matrices were limited to the protein-coding genes. Analyses of differentially expressed genes were performed using the DESeq2 package. During the analyses, multiple group comparisons were performed (A2 vs. A1; A3 vs. A2; A4 vs. A2; A5 vs. A2). Obtained results were filtered based on adjusted p-values (p≤0.01) and fold change (FC) [abs(logFC) <1] values. Finally, results were visualized using the EnhancedVolcano package. The analyses, filtration of the obtained results, and visualizations were performed by using R programming language.

Gene enrichment analysis. Using g: Profiler/g:Orth (https://biit.cs.ut.ee/gprofiler/orth) and Mouse Genome Informatics (https://www.informatics.jax.org/), each of the genes that initially emerged was converted from a mouse gene to a human gene. Enrichr (https://maayanlab.cloud/Enrichr/) was used to determine related pathways for genes affected in each of the three radiation treatment groups (A3 vs. A2; A4 vs. A2; A5 vs. A2). Additionally, Venn diagram analysis was employed to identify the genes shared by each group. The expression status of the shared genes detected in all groups was evaluated separately according to the amount of radiation and filtered according to linear increase or decrease in expression. In addition, the presence of transcription factor elements among the genes identified as common was determined using the TRRUST (https://www.grnpedia.org/trrust/) and STRING tools (https://string-db.org/).

Evaluation of expression differences between our murine data and TCGA. UALCAN (https://ualcan.path.uab.edu/), which utilizes The Cancer Genome Atlas (TCGA) datasets, was used as a comprehensive, accessible and dynamic online tool for OMICS data analysis in cancer for revealing expression changes in tumor compared to normal tissue (p <0.01). g:Profiler-converted genes (from mouse to human) were used for these analyses.

Correlation and survival analysis according to prominently altered genes. TIMER 2.0 analysis tool (http://timer.cistrome.org/) was used to evaluate correlation of gene expression, with partial Spearman’s rho <0 for negative correlation and rho >0 for positive correlation and p<0.05 accepted as significant. Furthermore, the Kaplan–Meier plotter (https://kmplot.com/analysis/) was used to determine the overall survival according to prominent genes. p-Values of less than 0.05 were considered significant.

Results

Differentially expressed genes. As a result of the analysis, 2,814 genes (1,301 up-regulated and 1,513 down-regulated) were differentially expressed in group A2 compared with A1; 31 genes (20 up-regulated and 11 down-regulated) in A3 vs. A2; 18 genes (14 up-regulated and 4 down-regulated) in A4 vs. A2; and 52 genes (26 up-regulated and 26 down-regulated) in A5 vs. A2. Their volcano plot analyses are shown in Figure 1.

Figure 1.
Figure 1.

Volcano plots of differentially expressed genes in murine data. The genes with the highest or lowest expression among the differentially expressed genes are indicated. Mice were without lung cancer (A1), with lung cancer but untreated (A2), treated with 20 Gy radiotherapy with flattening filter (FF) beam at 400 MU/min (A3), with flattening filter-free (FFF) beams at 1,000 (A4) and 1,800 (A5) MU/min dose rates. The mice were irradiated 20 days after A549 cancer cell-line implantation and sacrificed 48 h after irradiation for genetic analysis. Green dot: |log2FC| <1, Red dot: p<0.05 and |log2FC|<1; NS: Not significant.

Evaluation of enrichment analysis of differentially expressed genes. Pathway analyses for deregulated genes were performed separately for each group. According to these analyses, glutamergic synapse, actin cytoskeleton regulation, and steroid synthesis were the pathways significantly altered in A3 vs. A2 and A4 vs. A2 groups, whereas RNA transport, actin cytoskeleton regulation and phagosome-related pathways were significantly altered in the A5 group, in which the radiotherapy dose rate (1800 MU/min) was the highest, vs. the A2 group (Table I).

View this table:
Table I.

The pathway analysis of human counterparts of genes differentially expressed (DEGs) in mice with lung cancer treated with 20 Gy radiotherapy with flattening filter beam at 400 MU/min (A3), or with flattening filter-free beams at 1,000 (A4) or 1,800 (A5) MU/min dose rates compared with the untreated lung cancer control group.

In addition, genes shared by at least two of the radiation-treated groups were identified and evaluated together with the expression levels determined in the A2 vs. groups to reveal how they changed with the effects of radiation (Figure 2).

Figure 2.
Figure 2.

Venn diagram analysis of differentially expressed genes in study groups. Mice were without lung cancer (A1), with lung cancer but untreated (A2) or treated with 20 Gy radiotherapy with flattening filter (FF) beam at 400 MU/min (A3), or with flattening filter-free (FFF) beams at 1,000 (A4) or 1,800 (A5) MU/min dose rates murine data. A: Up/down-regulated genes for all groups. B: Up/down-regulated genes for groups receiving radiotherapy at different dose rates compared with the untreated lung cancer group (A2).

Accordingly, 12 genes [integrin subunit alpha M (Itgam), GPN-loop GTPase 1 (Gpn1), podocalyxin-like (Podxl), TBC1 domain family member 31 (Tbc1d31), lanosterol synthase (Lss), family with sequence similarity 83 member A (Fam83a), muskelin1 (Mklin1), integrin subunit alpha D (Itgad), armadillo repeat containing 5 (Armc5), armadillo repeat containing X-linked 3 (Armcx3), CCR4-NOT transcription complex subunit 4 (Cnot4), derlin 1 (Derl1)] were found to be shared by all radiotherapeutic groups (Table II). In addition, the expression of most of these genes were found to progressively increase or decrease as the dose rate increased. Among these 12 genes, seven were also shared by the A2 and A1 groups. While six of these shared genes were down-regulated in the untreated cancer group (A2) compared with the normal control, among them Mkln1, Podxl and Itgam were found to have increasing expression as the dose rate increased in the treated groups. On the other hand, of these shared genes, only Gpn1 showed increased expression in the untreated cancer group (A2) compared to the healthy group (A1), while it was found to have gradually decreasing expression in the treated groups (A3, A4 and A5) (Table II). Among the genes that were not detected as being differentially expressed in the A2 vs. A1 group but which were affected by radiotherapy were Cnot4, Tbc1d31, Armcx3, Fam83a and Derl1. All genes for which remarkable changes were observed in the datasets are listed in Table II. Moreover, Nfyb and Cnot4, among the prominent genes, were found to act as transcription factors/transcriptional regulator respectively by using TRRUST and STRING databases.

View this table:
Table II.

Genes shared among groups whose expression was significantly affected by radiation therapy. Mice were without lung cancer (A1), with lung cancer but untreated (A2) or treated with 20 Gy radiotherapy with flattening filter (FF) beam at 400 MU/min (A3), or with flattening filter-free (FFF) beams at 1,000 (A4) or 1,800 (A5) MU/min dose rates murine data.

Evaluation of our prominent DEGs compared with TGCA expression data. According to UALCAN, 12 out of 21 genes were found to be statistically significantly differently expressed between tumor and control tissues in TGCA data and only five (LSS, PODXL, GPN1, ITGAM, ATAD1) showed expression similarities with our data in mice (Figure 3).

Figure 3.
Figure 3.
Figure 3.

Comparison of gene expression profiles between human cancer tissues and key genes differentially expressed in radiotherapy-treated mice with non-small-cell lung cancer. Box plots display the expression levels of selected target genes based on data from The Cancer Genome Atlas (TCGA). *Only genes showing significant expression differences (p<0.01) similar to those observed in the mouse model are included. In the plots, red bars represent expression in primary tumor tissues (n=515), while blue bars represent expression in normal (healthy) tissues (n=59); n: Number of samples. AOAH: Acyloxyacyl hydrolase; ATAD1: ATPase family AAA domain containing 1; DERL1: derlin1; FAM83A: family with sequence similarity 83 member A; GPN1: GPN-loop GTPase 1; ITGAM: integrin subunit alpha M; LSS: lanosterol synthase; MKLN1: muskelin 1; MRM1: mitochondrial rRNA methyltransferase 1; NFYB: nuclear transcription factor Y subunit beta; PODXL: podocalyxin like; POLA1: DNA polymerase alpha 1, catalytic subunit.

Gene correlation and overall survival analysis according to expression of prominent genes. MKLN1 gene was found to positively correlate with CNOT4 and POLA1 according to TCGA data (cancer vs. control) using TIMER 2.0 tool, and this was also the case in our groups receiving treatment. In contrast, ITGAD and ITGAM, which were positively correlated with AOAH according to the TCGA dataset (Figure 4), showed negative correlation in our murine dataset. Although Mkln1 expression was increased in mice receiving treatment in our data, when compared to the human dataset, this gene was found to be highly expressed in individuals with cancer.

Figure 4.
Figure 4.

Correlation analysis of human genes matching our murine data using TIMER 2.0. AOAH: Acyloxyacyl hydrolase; CNOT4: CCR4-NOT transcription complex, subunit 4; ITGAD: integrin subunit alpha D; ITGAM: integrin subunit alpha M; Log2TPM: log base 2 of transcripts per million; MKLN1: muskelin 1; POLA1: DNA polymerase alpha 1, catalytic subunit.

In the analysis of overall survival using the TGCA dataset, low expression of ATAD1 and GPN1 genes was found to be associated with higher survival rates (Figure 5). In our murine data, these genes also had lower expression in treated groups (A3, A4 and A5) compared to the untreated cancer group (A2). In our treated groups, expression of Lss, Armc5, Itgad and Armcx3 genes were increased. Analysis of TGCA survival data revealed that higher levels of expression of these genes corresponded with higher survival (Figure 5).

Figure 5.
Figure 5.
Figure 5.
Figure 5.

Survival analysis according to expression of human genes that were found to be significant and matched in our murine data. AOAH: Acyloxyacyl hydrolase; ARMC5: armadillo repeat containing 5; ARMCX3: armadillo repeat containing X-linked 3; ATAD1: ATPase family AAA domain containing 1; DERL1: derlin1; DMAC2: distal membrane arm assembly component 2; FAM83A: family with sequence similarity 83 member A; GPN1: GPN-loop GTPase 1; HR: hazard ratio; ITGAD: integrin subunit alpha D; ITGAM: integrin subunit alpha M; LSS: lanosterol synthase; TBC1D31: TBC1 domain family member 31; TVP23A: trans-Golgi network vesicle protein 23 homolog A.

Discussion

It is predicted that a better and more accurate application of clinical radiotherapy treatment approach can be provided by understanding the molecular biology of the LCa treatment process. The success of radiotherapy in treating tumors such as NSCLC may be limited by genetic mutations or changes in the tumor microenvironment, leading to disease progression. The literature has not yet clarified the mechanisms, molecular events, and markers underlying radiotherapy response at the genetic level for NSCLC. Therefore, it is important to demonstrate the changes of gene expression in LCa radiotherapy with FF and FFF beams. The prediction of radiation response is a complex process that depends on both treatment factors and the tumor’s molecular profile. Therefore, NGS has potential as a tool to predict tumor radiation response. A combination of genetic variants and biological mechanisms are likely involved in determining radiation responses. This means that multiple genes and their interactions need to be considered to accurately predict tumor radiation response. Predicting tumor radiation response using NGS is challenging due to the complexity of the underlying biological mechanisms. The current literature has limited knowledge on these mechanisms, making it difficult to accurately predict radiation response using NGS (14).

In our study, different dose rates (FF vs. FFF) were evaluated in order to elucidate the effects of radiotherapy on molecular mechanisms and changes in tumor structure. The detection of changes in gene expression levels caused by FF and FFF beams in LCa was investigated in this study. According to our results, Cnot4 gene gradually increased in expression in the groups receiving therapy. In a study by Zhang et al., CNOT4 gene was found to have low expression in the A549 cell line compared to the control. Furthermore, CNOT4 overexpression was found to inhibit tumor cell proliferation, colony formation, cell migration and invasion, and increase cell apoptosis. In addition, CNOT4 overexpression enhanced cancer cell response to cytotoxicity mediated by CTLs (15). Furthermore, in a xenograft model, CNOT4 overexpression was reported to restrict tumor growth and enhance the effect of programmed death-ligand 1-targeted immunotherapy accompanied by greater T-lymphocyte infiltration and higher levels of lymphokines (16). Therefore, it is suggested that radiotherapy affects the expression of CNOT4 gene and this may cause cancer to shrink and show a positive reaction.

The NFYB gene, a transcription factor, is required for numerous biological processes such as apoptosis, senescence, and cell division. Studies have shown that NFYB can increase STK33 activity and promote cisplatin resistance in diffuse large B-cell lymphoma (17). NFYB induces high expression of E2F1 by enhancing the CHK1 signaling pathway and promotes oxaliplatin resistance in colorectal cancer (18). In our study, the NFYB gene was overexpressed in the untreated group, while its expression decreased in the treated groups (A3 and A5) compared to the untreated cancer group. This suggests that radiotherapy plays an active role in tumor shrinkage by decreasing the expression of this transcription factor.

In previous studies, it was reported that the ATAD1 gene was deleted at various rates (7-25%) in different cancer types, such as prostatic, melanoma, and glioblastoma, and the growth rate of tumors with ATAD1 gene deletion was significantly reduced. ATAD1 deficiency makes cells more sensitive to ubiquitin-proteasome system dysfunction, which predisposes the cells to apoptosis (19). In this case, loss of function of ATAD1 is a vulnerability for tumor cells. In our study, it appears that expression of Atad1 gradually decreased depending on treatment. GPN1, which has a critical role in DNA repair and transcription, is important for the nuclear localization and biogenesis of human RNA polymerase II. In the work conducted by Hernandez et al. (20), it was emphasized that GPN1 works by forming a complex with GPN3 in mammals and that GPN–GPN3 is important in the maturation and nuclear traffic of eukaryotic RNA pol II. Therefore, failure to express GPNs is lethal in eukaryotes (21). On the other hand, another study showed that GPN3 is necessary for the proliferation of breast cancer cells (22). The decrease in the expression of both ATAD1 and GPN1 due to treatment was evaluated as being adverse to tumor development.

In our study, Lss gene expression increased dose rate-dependently. Abnormal cholesterol metabolism by cancer cells is one of the pathways that can be used to limit tumor growth and the metastatic process. Therefore, inhibition of LSS can be identified as a therapeutic target. LSS inhibition has been shown to inhibit tumor proliferation (23). Furthermore, it is thought that cholesterol accumulation in tumor cells plays a role in the development of resistance. Hua et al. have shown that there is a link between increasing amounts of lanosterol and drug resistance. In a study conducted to understand the role of LSS, it was observed that HLE cells exposed to UV-B overexpressed LSS. Overexpression of LSS is caused by reactive oxygen species production (24). These findings may explain an increase in the amount of LSS expression in the radiotherapy groups in our study.

Although the ARMC5 gene has been shown to be a tumor suppressor, little is known about the mechanism of tumor formation due to ARMC5 loss (25). However, in our study, Armc5 gene was found to have an increased expression in the treated groups. Another gene that we thought important is ARMCX3. Studies showing that ARMCX3 is up-regulated due to high lipid amounts (26) suggest that ARMCX3 may increase in response to increasing LSS in our study. In addition, ALEX3 encoded by the ARMCX3 gene, has been shown to suppress the invasion of non-small cell LCa using the protein kinase B (AKT)/Snail family transcriptional repressor 2 (Slug/SNAI2) signaling pathway/E-cadherin signaling pathway (27). DERL1 expression is thought to increase due to radiation-induced endoplasmic reticulum stress (28). ITGAD plays a role in the regulation of the actin cytoskeleton (29) and in our study, it was observed that the ARMC5 gene together with ITGAD was up-regulated in radiotherapy treated groups.

Revealing the radiation response of a cancer can help develop strategies to individual tailor treatment to each patient. This can lead to more effective and specialized treatment plans in radiation oncology, improving treatment outcomes and reducing the risk of recurrence. By analyzing genomic data obtained through NGS, researchers can improve the ability to respond positively to radiotherapy and personalize radiation treatments for individual patients. Therefore, NGS is expected to play an increasingly important role in radiation oncology in the future. Although FFF beams have many dosimetric advantages, there is limited information available in the literature regarding their radiobiological effects on acute in vitro/in vivo cancer cells. Therefore, it is crucial to determine the exact genetic and radiobiological DNA damage after dose delivery at different dose rates in LCa radiotherapy, as the probability of tumor control increases with high doses of stereotactic radiosurgery and stereotactic body radiotherapy. Numerous in-vitro studies have been conducted to investigate whether dose rate is a significant dosimetric parameter for cell survival when applying FF and FFF beams to various cancer cell lines (30-32). However, there is limited knowledge regarding the potential effects of FFF beams on cancer cells, and previous in-vitro radiobiological studies have reported significant contradictions (33, 34). Uncertainty remains regarding the in-vivo radiobiological consequences of the FFF clinical dose rate effect. Therefore, it is important to investigate whether there is a difference in cell survival between FF and FFF irradiation. The present study observed significant changes in gene expression levels due to high FFF dose rates of 1000 and 1800 MU/min compared to the low dose rate of 400 MU/min FF. Therefore, it was concluded that FFF beams are more effective than FF beams in altering gene expression.

An important limitation of our study is that the comparison of 400 MU/min (FF) and 1,000-1,800 MU/min (FFF) groups included not only the difference in dose rate but also the physical differences specific to the beam type used. Parameters such as spectral structure, dose/pulse, profile homogeneity and scattering doses differ between FF and FFF beams. Therefore, the gene expression changes obtained cannot be attributed solely to the dose rate effect. Although our findings reflect realistic conditions in clinical applications, additional studies using the same beam type and different dose rates are needed in the future to clearly demonstrate the biological effects of pure dose rate differences.

In this study, we analyzed only the changes in gene expression levels of FF and FFF irradiation on LCa in the acute period in experimental nude mouse models. However, future studies may include genetic analyses for different types of cancer in addition to subacute and chronic effects.

Conclusion

Genes whose expression was significantly altered at different clinical dose rates were found to be involved in tumor proliferation, growth, migration, and invasion. Additionally, compared to FF beams, it was concluded that FFF beams caused more significant changes in gene expression levels in NSCLC. These genes require further confirmation through human/mice studies.

Footnotes

  • Authors’ Contributions

    Conceptualization: SA. Data curation: TKK and AK. Formal analysis, funding acquisition: SA, TKK. Investigation: SA, EE and EÇ. Methodology: SA, TKK and JB. Project administration: SA. Resources: TKK, BEÇ, JB and AK. Software: AK. Supervision: SA. Validation: SA. Visualization: TKK and AK. Writing - original draft: SA, TKK, BEÇ, EE, EÇ and AK. Writing - review and editing: SA, TKK, BEÇ, EE, EÇ and AK.

  • Conflicts of Interest

    The Authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. This manuscript has not been published before and it is not under consideration for publication anywhere else.

  • Funding

    The Authors extend their appreciation to Türkiye Health Institutes Presidency (TUSEB) for funding this research through project number (TÜSEB 2022-B-02, Project Number 22860).

  • Artificial Intelligence (AI) Disclosure

    During the preparation of this work the Authors used DeepL Translate for English translation. After using this tool/service, the Authors reviewed and edited the content as needed and take full responsibility for the content of the publication.

  • Received May 25, 2025.
  • Revision received June 26, 2025.
  • Accepted July 8, 2025.

This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).

References

  1. ↵
    1. Vinod SK,
    2. Hau E
    : Radiotherapy treatment for lung cancer: Current status and future directions. Respirology 25(S2): 61-71, 2020. DOI: 10.1111/resp.13870
    OpenUrlCrossRefPubMed
  2. ↵
    1. Yang WC,
    2. Hsu FM,
    3. Yang PC
    : Precision radiotherapy for non-small cell lung cancer. J Biomed Sci 27(1): 82, 2020. DOI: 10.1186/s12929-020-00676-5
    OpenUrlCrossRefPubMed
  3. ↵
    1. Gul OV
    : Experimental evaluation of out-of-field dose for different high-energy electron beams and applicators used in external beam radiotherapy. Radiat Phys Chem 215: 111345, 2024. DOI: 10.1016/j.radphyschem.2023.111345
    OpenUrlCrossRef
  4. ↵
    1. Gul OV,
    2. Duzova M
    : Effect of different CTV shrinkage and skin flash margins on skin dose for left chest wall IMRT: A dosimetric study. Radiat Phys Chem 216: 111445, 2024. DOI: 10.1016/j.radphyschem.2023.111445
    OpenUrlCrossRef
  5. ↵
    1. Xiao Y,
    2. Zhuang H
    : Effect of stereotactic radiotherapy on immune microenvironment of lung cancer. Front Immunol 13: 1025872, 2022. DOI: 10.3389/fimmu.2022.1025872
    OpenUrlCrossRefPubMed
  6. ↵
    1. Pokhrel D,
    2. Halfman M,
    3. Sanford L
    : FFF-VMAT for SBRT of lung lesions: Improves dose coverage at tumor-lung interface compared to flattened beams. J Appl Clin Med Phys 21(1): 26-35, 2020. DOI: 10.1002/acm2.12764
    OpenUrlCrossRef
  7. ↵
    1. Aoki S,
    2. Yamashita H,
    3. Haga A,
    4. Nawa K,
    5. Imae T,
    6. Takahashi W,
    7. Abe O,
    8. Nakagawa K
    : Flattening filter-free technique in volumetric modulated arc therapy for lung stereotactic body radiotherapy: A clinical comparison with the flattening filter technique. Oncol Lett 15(3): 3928-3936, 2018. DOI: 10.3892/ol.2018.7809
    OpenUrlCrossRefPubMed
  8. ↵
    1. Bezjak A,
    2. Paulus R,
    3. Gaspar LE,
    4. Timmerman RD,
    5. Straube WL,
    6. Ryan WF,
    7. Garces YI,
    8. Pu AT,
    9. Singh AK,
    10. Videtic GM,
    11. McGarry RC,
    12. Iyengar P,
    13. Pantarotto JR,
    14. Urbanic JJ,
    15. Sun AY,
    16. Daly ME,
    17. Grills IS,
    18. Sperduto P,
    19. Normolle DP,
    20. Bradley JD,
    21. Choy H
    : Safety and efficacy of a five-fraction stereotactic body radiotherapy schedule for centrally located non-small-cell lung cancer: NRG Oncology/RTOG 0813 trial. J Clin Oncol 37(15): 1316-1325, 2019. DOI: 10.1200/JCO.18.00622
    OpenUrlCrossRefPubMed
  9. ↵
    1. Kanyilmaz G,
    2. Aktan M,
    3. Koc M,
    4. Demir H,
    5. Demir LS
    : Unplanned irradiation of internal mammary lymph nodes in breast cancer. Radiol Med 122(6): 405-411, 2017. DOI: 10.1007/s11547-017-0747-5
    OpenUrlCrossRefPubMed
  10. ↵
    1. Bibault JE,
    2. Tinhofer I
    : The role of Next-Generation Sequencing in tumoral radiosensitivity prediction. Clin Transl Radiat Oncol 3: 16-20, 2017. DOI: 10.1016/j.ctro.2017.03.002
    OpenUrlCrossRefPubMed
  11. ↵
    1. Sabour L,
    2. Sabour M,
    3. Ghorbian S
    : Clinical applications of Next-Generation Sequencing in cancer diagnosis. Pathol Oncol Res 23(2): 225-234, 2017. DOI: 10.1007/s12253-016-0124-z
    OpenUrlCrossRefPubMed
  12. ↵
    1. Bleaney CW,
    2. Abdelaal H,
    3. Reardon M,
    4. Anandadas C,
    5. Hoskin P,
    6. Choudhury A,
    7. Forker L
    : Clinical biomarkers of tumour radiosensitivity and predicting benefit from radiotherapy: a systematic review. Cancers (Basel) 16(10): 1942, 2024. DOI: 10.3390/cancers16101942
    OpenUrlCrossRefPubMed
  13. ↵
    1. Navarria P,
    2. Ascolese AM,
    3. Mancosu P,
    4. Alongi F,
    5. Clerici E,
    6. Tozzi A,
    7. Iftode C,
    8. Reggiori G,
    9. Tomatis S,
    10. Infante M,
    11. Alloisio M,
    12. Testori A,
    13. Fogliata A,
    14. Cozzi L,
    15. Morenghi E,
    16. Scorsetti M
    : Volumetric modulated arc therapy with flattening filter free (FFF) beams for stereotactic body radiation therapy (SBRT) in patients with medically inoperable early stage non small cell lung cancer (NSCLC). Radiother Oncol 107(3): 414-418, 2013. DOI: 10.1016/j.radonc.2013.04.016
    OpenUrlCrossRefPubMed
  14. ↵
    1. Luo BH,
    2. Huang JQ,
    3. Huang CY,
    4. Tian P,
    5. Chen AZ,
    6. Wu WH,
    7. Ma XM,
    8. Yuan YX,
    9. Yu L
    : Screening of lymphoma radiotherapy-resistant genes with CRISPR activation library. Pharmgenomics Pers Med 16: 67-80, 2023. DOI: 10.2147/PGPM.S386085
    OpenUrlCrossRefPubMed
  15. ↵
    1. Zhang B,
    2. Chen S
    : CNOT4 suppresses non-small cell lung cancer progression and is required for effector cytolytic T lymphocytes cell responses to lung cancer cells. Mol Immunol 132: 165-171, 2021. DOI: 10.1016/j.molimm.2020.12.020
    OpenUrlCrossRefPubMed
  16. ↵
    1. Zhang B,
    2. Zhao B,
    3. Han S,
    4. Chen S
    : CNOT4 suppresses nonsmall cell lung cancer progression by promoting the degradation of PAF1. Mol Carcinog 62(10): 1563-1571, 2023. DOI: 10.1002/mc.23599
    OpenUrlCrossRefPubMed
  17. ↵
    1. Feng L,
    2. Xu X,
    3. Zhao K
    : NFYB potentiates STK33 activation to promote cisplatin resistance in diffuse large B-cell lymphoma. Leuk Res 111: 106708, 2021. DOI: 10.1016/j.leukres.2021.106708
    OpenUrlCrossRefPubMed
  18. ↵
    1. Fang Z,
    2. Gong C,
    3. Yu S,
    4. Zhou W,
    5. Hassan W,
    6. Li H,
    7. Wang X,
    8. Hu Y,
    9. Gu K,
    10. Chen X,
    11. Hong B,
    12. Bao Y,
    13. Chen X,
    14. Zhang X,
    15. Liu H
    : NFYB-induced high expression of E2F1 contributes to oxaliplatin resistance in colorectal cancer via the enhancement of CHK1 signaling. Cancer Lett 415: 58-72, 2018. DOI: 10.1016/j.canlet.2017.11.040
    OpenUrlCrossRefPubMed
  19. ↵
    1. Winter JM,
    2. Fresenius HL,
    3. Cunningham CN,
    4. Wei P,
    5. Keys HR,
    6. Berg J,
    7. Bott A,
    8. Yadav T,
    9. Ryan J,
    10. Sirohi D,
    11. Tripp SR,
    12. Barta P,
    13. Agarwal N,
    14. Letai A,
    15. Sabatini DM,
    16. Wohlever ML,
    17. Rutter J
    : Collateral deletion of the mitochondrial AAA+ ATPase ATAD1 sensitizes cancer cells to proteasome dysfunction. Elife 11: e82860, 2022. DOI: 10.7554/eLife.82860
    OpenUrlCrossRefPubMed
  20. ↵
    1. Méndez-Hernández LE,
    2. Pérez-Mejía AE,
    3. Lara-Chacón B,
    4. Barbosa-Camacho AA,
    5. Peña-Gómez SG,
    6. Martínez-Sánchez M,
    7. Robledo-Rivera AY,
    8. Sánchez-Olea R,
    9. Calera MR
    : Gpn1 and Gpn3 associate tightly and their protein levels are mutually dependent in mammalian cells. FEBS Lett 588(21): 3823-3829, 2014. DOI: 10.1016/j.febslet.2014.08.038
    OpenUrlCrossRefPubMed
  21. ↵
    1. Korf L,
    2. Ye X,
    3. Vogt MS,
    4. Steinchen W,
    5. Watad M,
    6. van der Does C,
    7. Tourte M,
    8. Sivabalasarma S,
    9. Albers S-V,
    10. Essen L-O
    : Archaeal GPN-loop GTPases involve a lock-switch-rock mechanism for GTP hydrolysis. mBio 14(6): e0085923, 2023. DOI: 10.1128/mbio.00859-23
    OpenUrlCrossRefPubMed
  22. ↵
    1. Lara-Chacón B,
    2. Guerrero-Rodríguez SL,
    3. Ramírez-Hernández KJ,
    4. Robledo-Rivera AY,
    5. Velazquez MAV,
    6. Sánchez-Olea R,
    7. Calera MR
    : Gpn3 is essential for cell proliferation of breast cancer cells independent of their malignancy degree. Technol Cancer Res Treat 18: 1533033819870823, 2019. DOI: 10.1177/1533033819870823
    OpenUrlCrossRefPubMed
  23. ↵
    1. Sun X,
    2. Zhang J,
    3. Liu H,
    4. Li M,
    5. Liu L,
    6. Yang Z,
    7. Hu W,
    8. Bai H,
    9. Xu J,
    10. Xing J,
    11. Xu Z,
    12. Mo A,
    13. Guo Z,
    14. Bai Y,
    15. Zhou Q,
    16. Wang Y,
    17. Zhang S,
    18. Zhang S
    : Lanosterol synthase loss of function decreases the malignant phenotypes of HepG2 cells by deactivating the Src/MAPK signaling pathway. Oncol Lett 26(1): 295, 2023. DOI: 10.3892/ol.2023.13881
    OpenUrlCrossRefPubMed
  24. ↵
    1. Hua H,
    2. Yang T,
    3. Huang L,
    4. Chen R,
    5. Li M,
    6. Zou Z,
    7. Wang N,
    8. Yang D,
    9. Liu Y
    : Protective effects of lanosterol synthase up-regulation in UV-B-induced oxidative stress. Front Pharmacol 10: 947, 2019. DOI: 10.3389/fphar.2019.00947
    OpenUrlCrossRefPubMed
  25. ↵
    1. Correa R,
    2. Zilbermint M,
    3. Berthon A,
    4. Espiard S,
    5. Batsis M,
    6. Papadakis GZ,
    7. Xekouki P,
    8. Lodish MB,
    9. Bertherat J,
    10. Faucz FR,
    11. Stratakis CA
    : The ARMC5 gene shows extensive genetic variance in primary macronodular adrenocortical hyperplasia. Eur J Endocrinol 173(4): 435-440, 2015. DOI: 10.1530/EJE-15-0205
    OpenUrlAbstract/FREE Full Text
  26. ↵
    1. Mirra S,
    2. Gavaldà-Navarro A,
    3. Manso Y,
    4. Higuera M,
    5. Serrat R,
    6. Salcedo MT,
    7. Burgaya F,
    8. Balibrea JM,
    9. Santamaría E,
    10. Uriarte I,
    11. Berasain C,
    12. Avila MA,
    13. Mínguez B,
    14. Soriano E,
    15. Villarroya F
    : ARMCX3 mediates susceptibility to hepatic tumorigenesis promoted by dietary lipotoxicity. Cancers (Basel) 13(5): 1110, 2021. DOI: 10.3390/cancers13051110
    OpenUrlCrossRefPubMed
  27. ↵
    1. Du J,
    2. Zhang X,
    3. Zhou H,
    4. Miao Y,
    5. Han Y,
    6. Han Q,
    7. Wang E
    : Alex3 suppresses non–small cell lung cancer invasion via AKT/Slug/E-cadherin pathway. Tumour Biol 39(7): 101042831770144, 2017. DOI: 10.1177/1010428317701441
    OpenUrlCrossRef
  28. ↵
    1. Kim W,
    2. Lee S,
    3. Seo D,
    4. Kim D,
    5. Kim K,
    6. Kim E,
    7. Kang J,
    8. Seong KM,
    9. Youn H,
    10. Youn B
    : Cellular stress responses in radiotherapy. Cells 8(9): 1105, 2019. DOI: 10.3390/cells8091105
    OpenUrlCrossRef
  29. ↵
    1. Wen H,
    2. Kim YC,
    3. Snyder C,
    4. Xiao F,
    5. Fleissner EA,
    6. Becirovic D,
    7. Luo J,
    8. Downs B,
    9. Sherman S,
    10. Cowan KH,
    11. Lynch HT,
    12. Wang SM
    : Family-specific, novel, deleterious germline variants provide a rich resource to identify genetic predispositions for BRCAx familial breast cancer. BMC Cancer 14: 470, 2014. DOI: 10.1186/1471-2407-14-470
    OpenUrlCrossRefPubMed
  30. ↵
    1. Aras S,
    2. Ozkanli S,
    3. Erdem E,
    4. Gokalp S,
    5. Erdogan CE
    : Investigation of low and high dose rate X-ray effects on histopathological changes and prognostic importance of Ki-67 in laryngeal cancer radiotherapy. Appl Radiat Isot 197: 110823, 2023. DOI: 10.1016/j.apradiso.2023.110823
    OpenUrlCrossRef
    1. Nakano H,
    2. Minami K,
    3. Yagi M,
    4. Imaizumi H,
    5. Otani Y,
    6. Inoue S,
    7. Takashina M,
    8. Seo Y,
    9. Takahashi Y,
    10. Sumida I,
    11. Ogawa K,
    12. Koizumi M
    : Radiobiological effects of flattening filter-free photon beams on A549 non-small-cell lung cancer cells. J Radiat Res 59(4): 442-445, 2018. DOI: 10.1093/jrr/rry041
    OpenUrlCrossRefPubMed
  32. ↵
    1. Verbakel WF,
    2. van den Berg J,
    3. Slotman BJ,
    4. Sminia P
    : Comparable cell survival between high dose rate flattening filter free and conventional dose rate irradiation. Acta Oncol 52(3): 652-657, 2013. DOI: 10.3109/0284186X.2012.737021
    OpenUrlCrossRefPubMed
  33. ↵
    1. Lohse I,
    2. Lang S,
    3. Hrbacek J,
    4. Scheidegger S,
    5. Bodis S,
    6. MacEdo NS,
    7. Feng J,
    8. Lütolf UM,
    9. Zaugg K
    : Effect of high dose per pulse flattening filter-free beams on cancer cell survival. Radiother Oncol 101(1): 226-232, 2011. DOI: 10.1016/j.radonc.2011.05.072
    OpenUrlCrossRefPubMed
  34. ↵
    1. Sarojini S,
    2. Pecora A,
    3. Milinovikj N,
    4. Barbiere J,
    5. Gupta S,
    6. Hussain ZM,
    7. Tuna M,
    8. Jiang J,
    9. Adrianzen L,
    10. Jun J,
    11. Catello L,
    12. Sanchez D,
    13. Agarwal N,
    14. Jeong S,
    15. Jin Y,
    16. Remache Y,
    17. Goy A,
    18. Ndlovu A,
    19. Ingenito A,
    20. Suh KS
    : A combination of high dose rate (10X FFF/2400 MU/min/10 MV X-rays) and total low dose (0.5 Gy) induces a higher rate of apoptosis in melanoma cells in vitro and superior preservation of normal melanocytes. Melanoma Res 25(5): 376-389, 2015. DOI: 10.1097/CMR.0000000000000174
    OpenUrlCrossRefPubMed
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools
Reprints and Permissions
Transcriptome Analysis of the Effects of X-Ray Radiotherapy on Non-small-cell Lung Cancer Using Next-generation Sequencing
SERHAT ARAS, TUĞBA KUL KÖPRÜLÜ, BURÇIN ERKAL ÇAM, JÜLIDE BALKAN, ESRA ERDEM, ESRA ÇIKLER, ALTAN KARA
In Vivo Sep 2025, 39 (5) 2711-2727; DOI: 10.21873/invivo.14070
Twitter logo Facebook logo Mendeley logo

Jump to section

Keywords