SPOCK1 acts as a negative regulator of ferroptosis in non-small cell lung cancer

Na Liu; Xiao-Ren Zhu; Yang Liu; Guo-Hu Han; Yuan-Yuan Liu; Kai-Xin Zhou; Min-Bin Chen

doi:10.1038/s41698-026-01455-5

. 2026 May 14;10:370. doi: 10.1038/s41698-026-01455-5

SPOCK1 acts as a negative regulator of ferroptosis in non-small cell lung cancer

Na Liu ^1,^2,^3,^✉,^#, Xiao-Ren Zhu ^1,^2,^3,^#, Yang Liu ^4,^#, Guo-Hu Han ⁵, Yuan-Yuan Liu ², Kai-Xin Zhou ⁶, Min-Bin Chen ^1,^✉

PMCID: PMC13639079 PMID: 42135411

Abstract

Non-small cell lung cancer (NSCLC) is a leading cause of cancer-related morbidity and mortality worldwide. Therefore, there is an urgent need to discover precise diagnostic markers and novel therapeutic strategies. A genome-wide association study (GWAS) was conducted using the Kunshan Elderly Cohort to identify genetic susceptibility factors underlying NSCLC. GWAS results identified that seven single-nucleotide polymorphisms (SNPs) were associated with NSCLC risk, including SPOCK1 (SPARC/osteonectin, CWCV, and Kazal-like domains proteoglycan (1), which emerged as a key candidate oncogene driving NSCLC progression. High expression of SPOCK1 in NSCLC tissues was associated with poor prognosis. Subsequently, using NSCLC cell lines with shRNA-mediated SPOCK1 knockdown (loss of function) and plasmid-mediated SPOCK1 overexpression (gain of function), we demonstrated that SPOCK1 significantly promotes in vitro NSCLC cell proliferation, migration, invasion, and cell cycle progression. Furthermore, in vivo mouse studies demonstrated that xenografted SPOCK1-knockdown NSCLC cells exhibited significantly reduced tumor growth and metastasis, whereas xenografted SPOCK1-overexpressing NSCLC cells enhanced tumor progression and metastasis. Mechanistically, SPOCK1 activates the JAK2/STAT3 signaling pathway by interacting with integrin alpha5 beta1 (α5β1) and subsequently upregulating SLC3A2 expression. This regulatory cascade promotes glutathione (GSH) biosynthesis and mitigates the accumulation of reactive oxygen species (ROS) and lipid peroxidation LPO), thereby suppressing ferroptosis in the NSCLC cells. Furthermore, Heat Shock Factor 1 (HSF1) is a critical upstream transcriptional regulator of SPOCK1. Notably, SPOCK1 acts as a pivotal mediator of the immunosuppressive microenvironment in the NSCLC tumors by inducing T-cell exhaustion and promoting infiltration of regulatory T cells (Tregs). In summary, SPOCK1 drives NSCLC progression by inhibiting ferroptosis and inducing an immunosuppressive microenvironment. Therefore, SPOCK1 may be a promising therapeutic target in NSCLC.

Subject terms: Cancer, Cell biology, Oncology

Introduction

Non-small cell lung cancer (NSCLC) accounts for ~85% of all lung cancer cases and is a leading cause of cancer-related mortality worldwide^1,2. The discovery and application of targeted therapies (e.g., EGFR/ALK inhibitors) and immune checkpoint blockade therapies (PD-1/PD-L1) have significantly improved NSCLC treatment^3–6. However, the 5-year survival rate for metastatic disease is ~4% because of an intrinsic lack of therapeutic targets and acquired resistance^3–6. Consequently, there is an urgent need to identify novel driver genes and clarify the mechanisms underlying therapy resistance for improving clinical outcomes.

To determine genetic susceptibility factors for NSCLC in the Chinese population, we conducted a genome-wide association study (GWAS) involving 380 patients and 3800 healthy controls and identified SPOCK1 (Testican-1) as a prominent susceptibility candidate. SPOCK1 is a matricellular glycoprotein of the SPARC/osteonectin family regulating cell adhesion and signaling^7,8. High expression of SPOCK1 is implicated in tumorigenesis, epithelial-mesenchymal transition, and drug resistance in malignancies such as glioblastoma and colorectal cancer^9–13. However, the specific function of SPOCK1 in NSCLC pathogenesis and its potential role in regulating metabolic changes are not well characterized.

Ferroptosis is an iron-dependent form of regulated cell death driven by LPO. It has emerged as a promising strategy to eliminate therapy-resistant tumors^14,15. However, cancer cells frequently evade ferroptosis by upregulating antioxidant defense systems such as the system Xc⁻ transporter^16–18. This heterodimeric complex is composed of SLC7A11 (a catalytic subunit) and SLC3A2 (a chaperone subunit) and mediates uptake of cystine for glutathione (GSH) biosynthesis, which is essential for GPX4-dependent neutralization of lipid peroxides^19–22. SLC7A11 is a well-established regulator of ferroptosis resistance. However, the upstream signaling networks that coordinate the stability and function of the SLC3A2 subunit in NSCLC remain poorly defined.

This study investigated the role and mechanisms by which SPOCK1 regulates NSCLC progression, including its regulatory role in SLC3A2 expression, ferroptosis, and tumor immune environment (TME).

Methods

Reagents and antibodies

The following primary antibodies were used in this study: anti-SPOCK1 (1:2000; ab229935, Abcam), anti-GPX4 (1:2000; BM5231, BOSTER), anti-GCLM (1:2000; M02948-2, BOSTER), anti-GCLC (1:2000; BM5017, BOSTER), anti-SLC7A11 (1:2000; BM5318, BOSTER), anti-SLC3A2 (1:2000; BM5641, BOSTER), anti-HSF1 (1:2000; M00250-1, BOSTER), anti-Ki67 (1:4000; ab16667, Abcam), anti-4HNE (1:4000; ab485067, Abcam), anti-GAPDH (1:4000; ab8245, Abcam), anti-JAK2 (1:2000; #3230, Cell Signaling Technology), anti-phospho-JAK2 (Tyr1007/1008) (1:1000; #3776, Cell Signaling Technology), anti-STAT3 (1:2000; #12640, Cell Signaling Technology), and anti-phospho-STAT3 (Tyr705) (1:2000; #9145, Cell Signaling Technology), anti-Integrin α5 (1:2000; #YM8809, Immunoway), anti-Integrin β1(1:2000; #YM8338, Immunoway), anti-pFAK (1:2000; #YM8684, Immunoway), anti-pSrc (1:2000; #YM8471, Immunoway). Ferrostatin-1 (Fer-1), Z-VAD-FMK (Z-VAD(OH)-FMK), Necrostatin-1 (Nec-1), 3-Methyladenine (3-MA), Erastin, and Stattic were purchased from MedChemExpress (MCE, Shanghai, China). All other reagents were purchased from Thermo Fisher Invitrogen (Suzhou, China).

Genome-wide association study

GWAS was conducted using a mixed linear model in fastGWA. Single-nucleotide polymorphisms (SNPs) with minor allele frequency (MAF) < 0.01 and insertion-deletion polymorphisms (indels) were excluded. Population structure confounding was addressed using PCA-adjusted (10 PCs) relatedness correction and age/sex covariates. Genome-wide significance was defined as P ≤ 5 × 10⁻⁸ (Bonferroni-corrected), and P ≤ 5 × 10⁻⁶ suggested potential association. GRCh38 (Human Genome version 38) reference genome was used for mapping genomic regions. MAF data were derived from the 1000 Genomes Project. Manhattan/Q-Q plots were generated in R 4.2.

Cell culture, clinical specimens, and tissue microarray

Human immortalized NSCLC cell lines (A549, H1650, H226, and H460) and normal human primary alveolar epithelial cells (HPAEpic) were obtained from the Shanghai Zhong Qiao Xin Zhou Biotechnology Co., Ltd. HPAEpiC cells were cultured in DMEM containing 10% FBS at 37 °C in 5% CO₂. A549 cells were grown in F12K medium containing 10% FBS. H460, H1650, and H226 cells were cultured in RPMI-1640 medium supplemented with 10% FBS. Primary human NSCLC cells (priNSCLC) were isolated and cultured as described previously^23–25. Fresh NSCLC and matched non-tumorous tissues were collected from primary NSCLC patients at the Affiliated Kunshan Hospital of Jiangsu University. Informed consent was obtained in accordance with the Declaration of Helsinki. All procedures were approved by the Ethics Committee of The Affiliated Kunshan Hospital of Jiangsu University (Approval No. 2025-03-018-H00-K01). A human NSCLC tissue microarray with 90 paired cases of NSCLC and matched adjacent non-tumorous tissues was obtained from Shanghai Outdo Biotech Co., Ltd. (Shanghai, China).

Single-cell RNA sequencing data and cell–cell communication analysis

Single-cell RNA sequencing (scRNA-seq) was performed on 14 samples (seven NSCLC tumor samples and seven adjacent normal tissue samples) from seven patients diagnosed with NSCLC. The scRNA-seq and cell-cell communication analysis were performed as described in previously published articles²⁶. All procedures were approved by the Ethics Committee of The Affiliated Kunshan Hospital of Jiangsu University (Approval No. 2025-03-018-H00-K01).

Untargeted metabolomics

Metabolomics profiling was performed using a UPLC-ESI-Q-Orbitrap-MS system (UHPLC, Shimadzu Nexera X2 LC-30AD, Shimadzu, Japan) coupled with Q-Exactive Plus (Thermo Scientific, San Jose, USA). Raw data were processed using the MSDIAL software for peak alignment, retention time correction, and peak area extraction. Identification of the metabolite structure was performed by precise mass matching (mass tolerance <10 ppm) and MS/MS spectral matching (mass tolerance <0.01 Da), as well as cross-referencing public databases, including HMDB, MassBank, and GNPS, and the in-house Bioprofile metabolite database (BP-DB). For data pre-processing, ion peaks with missing values >50% within groups were removed to exclude noise. Positive and negative ion data were normalized by Total Peak Area, integrated, and subjected to pattern recognition analysis using Python. All data underwent unit variance scaling prior to multivariate statistical analysis.

Multivariate statistical analysis was performed to evaluate metabolic profiles. PriNcipal Component Analysis (PCA) was used to visualize the overall distribution of samples and detect potential outliers. Subsequently, orthogonal partial least squares discriminant analysis (OPLS-DA) was used to maximize the separation between groups and identify discriminatory metabolic features. Differential metabolites were screened based on a combination of multivariate and univariate criteria, and features with a Variable Importance in the Projection value > 1 and a P value < 0.05 were considered statistically significant. A detailed workflow of the untargeted metabolomics approach is provided in the Supplementary Materials.

Plasmids, siRNAs, lentiviral construction, and cell transfection

The plasmids and siRNAs for this study were designed and synthesized by GenePharma Company (Shanghai, China). The pcDNA3.1-SLC3A2 overexpression construct or the pcDNA3.1 empty vector and the siRNAs were transfected into cells Lipofectamine 2000 (Invitrogen) for 48 h. Table S2 provides the detailed sequences of the siRNA.

The SPOCK1 shRNAs (SPOCK1-sh-s1: GCTTTCGAGACGATGATTATT and SPOCK1-sh-s2: CCCTTCAGAGATCAATGCCAT) and the SPOCK1-expressing cDNA sequence were custom-synthesized by Genechem (Shanghai, China). These sequences were individually cloned into the GV369 vector. Lentiviral particles were generated by co-transfecting the vector and lentivirus packaging plasmids into HEK-293 cells. Subsequently, the lentivirus was collected, filtered, and transduced into cultured NSCLC cells. Stable cell clones were selected in media containing 2 μg/mL puromycin. The stable SPOCK1-knockdown (SPOCK1-sh) or SPOCK1-overexpressing (OE-SPOCK1) cell lines were confirmed by qRT-PCR and western blotting. The shRNA sequences are listed in Table S2.

Cellular functional studies and gene/protein detection

Specific protein and mRNA expression levels were analyzed by western blotting and quantitative reverse transcription PCR (qRT-PCR), respectively. NSCLC cell growth and proliferation were assessed using the Cell Counting Kit-8 (CCK-8) and EdU (5-ethynyl-2’-deoxyuridine)/DAPI double labeling assays. Cell migration and invasion were analyzed using Transwell and Matrigel Transwell assays. These assays were conducted as described in previous studies^24,25,27,28.

Immunofluorescence and immunohistochemical staining

NSCLC cells were cultured overnight in laser confocal dishes. Then, the cells were washed three times with PBS and fixed in 4% paraformaldehyde for 20–30 min. They were permeabilized using 0.5% Triton X-100 for 5–10 min. The cells were blocked with BSA for 1 h at room temperature (RT) to prevent nonspecific binding. Then, they were incubated with primary antibodies overnight at 4 °C, followed by incubation with fluorescence-conjugated secondary antibodies for 2 h in darkness at RT. Nuclei were stained with DAPI for 10 min. The stained samples were then mounted and visualized using a confocal microscope (ZEISS, Oberkochen, Germany).

Immunohistochemical (IHC) staining was performed according to previously established methods^27,28. Tissues were fixed with 4% paraformaldehyde, paraffin embedded, and sectioned. The sections were then deparaffinized and rehydrated using xylene and graded ethanol. Antigen retrieval was performed via steam treatment in Ethylenediaminetetraacetic acid for 15 min, followed by a 10-min incubation in 3% H₂O₂. The sections were blocked with goat serum for 1 h at RT, followed by incubation with the primary antibodies overnight at 4 °C. This was followed by a 60-min incubation with the secondary antibody. Finally, immunostaining was performed using diaminobenzidine, and sections were counterstained with hematoxylin. The sections were observed and photographed under a light microscope.

Multiplex immunohistochemistry

The tissue microarrays (TMAs) for NSCLC were provided by the Shanghai Outdo Biotech Co., Ltd., Shanghai, China. The mIHC staining was performed using the Multiplex Immunohistochemistry (mIHC) kit (Akoya, NEL861001KT) according to the manufacturer’s instructions. Briefly, TMA sections were baked at 65 °C for 2 hours, followed by sequential dewaxing in an eco-friendly dewaxing solution (15 minutes × 2 cycles), absolute ethanol (5 minutes × 2 cycles), 90% ethanol (5 minutes), 75% ethanol (5 minutes), and distilled water (5 minutes). Antigen retrieval was performed by heating the samples in 200 mL of retrieval buffer in a microwave-safe container on high power for 3 minutes until boiling, followed by cooling to eliminate bubbles. The dewaxed sections were incubated in the buffer at low power for 20 minutes. The slides were washed with TBST, and excess liquid was blotted with tissue paper. A hydrophobic barrier was drawn around the tissue using an immunohistochemistry pen to prevent reagent diffusion. Blocking solution was applied within the barrier for 30 minutes, followed by incubation with diluted primary antibody at RT for 2 h. After washing the unbound primary antibody with TBST (5 minutes × 3 cycles), the samples were incubated with the species-matched secondary antibody at RT for 10 minutes. Subsequently, the samples were washed with TBST followed by incubation with fluorescent dye (1:100 dilution) for 15 minutes at RT. Residual dye was removed with TBST, and nuclei were stained with the DAPI working solution for 10 minutes. Excess DAPI was washed off, and the hydrophobic barrier was erased using a cotton swab dipped in eco-friendly dewaxing solution or xylene to avoid imaging artifacts. Residual solvents or debris were cleared with TBST. Finally, excess TBST around the tissue was blotted, and slides were mounted with coverslips using an appropriate mounting medium.

In vitro ferroptosis marker assays

Cellular levels of malondialdehyde (MDA), GSH disulfide (GSSG), and GSH were quantified using the MDA Test Kit (ADS-W-YH002, AIDISHENG, China), GSSG Test Kit (ADS-W-G002, AIDISHENG, China), and GSH Assay Kit (ADS-W-G001, AIDISHENG, China), respectively, in accordance with the manufacturer’s protocols. LPO and intracellular Fe²⁺ levels were assessed using the C11 BODIPY 581/591 molecular probe and Ferro Orange probe (F374, Dojindo, Japan), respectively, according to the manufacturer’s protocols. Briefly, cells were seeded in laser confocal dishes and cultured overnight or treated with or without Fer-1 for 24 h. The initial medium was removed and replaced with 1 mL of medium containing 5 μM C11 BODIPY 581/591 dye, followed by a 30-min incubation. Thereafter, cells were washed thrice with PBS and analyzed using a confocal fluorescence microscope. Intracellular Fe²⁺ concentrations were determined using the Ferro Orange probe.

For evaluating the levels of reactive oxygen species (ROS), cells were treated with 10 μM DCFH-DA in the original medium at 37 °C for 30 min. Then, the cells were washed three times with PBS and analyzed by fluorescence microscopy.

Mitochondrial superoxide levels were detected using the Mito-SOX assay kit. Briefly, cells were seeded in laser confocal dishes and incubated with 5 µM Mito-SOX™ reagent solution for 30 min. Then, the cells were washed three times with PBS, and mitochondrial ROS levels were assessed using a confocal microscope with excitation/emission wavelengths of 510/580 nm.

Transmission electron microscopy (TEM) assays

Cells were trypsinized and centrifuged at 1200 rpm for 5 min. Then, they were fixed with 2.5% glutaraldehyde at 4 °C for 4 h. The samples were subsequently washed with PBS thrice for 15 min each. Then, post-fixation was performed using 1% osmium tetroxide at 4 °C for 1 h. The cells were then dehydrated through a graded series of alcohol and acetone and embedded in Epon 812 resin. Polymerization was performed at 60 °C for 48 h. Ultra-thin slices (60–80 nm) were prepared using a Leica ultramicrotome. TEM images were captured using a JEM-1400Plus transmission electron microscope.

Co-immunoprecipitation (Co-IP)

Briefly, A549 cells were lysed in ice-cold IP lysis buffer supplemented with protease and phosphatase inhibitors. After centrifugation to remove cellular debris, the supernatants were incubated with an anti-SPOCK1 antibody or a normal IgG control overnight at 4 °C with gentle rotation. Subsequently, the samples were incubated with Protein A/G magnetic beads for an additional 4 h at 4 °C. The beads were washed three times with cold wash buffer, and the protein complexes were eluted by boiling in SDS loading buffer for 5 minutes. Finally, SPOCK1 and its interacting proteins were detected by western blotting.

Chromatin immunoprecipitation

Chromatin immunoprecipitation (ChIP) assay was performed using the SimpleChIP® Enzymatic Chromatin IP Kit (CST #9003) according to the manufacturer’s protocols. Chromatin extracts were immunoprecipitated with 10 µg monoclonal anti-phospho-STAT3 antibody or 1 µg normal IgG as a control. PCR primer sequences are listed in Table S2 and detailed methodologies are outlined in our previous study²⁴.

Dual-luciferase reporter assays

Luciferase reporter constructs were generated using pGL4-basic vectors containing segments of the human wild-type (WT) or mutant SLC3A2 promoter. 293 T cells were transiently transfected with the STAT3 pcDNA3.1 plasmid (Addgene, China), the Renilla luciferase reporter vector (RL-TK), and appropriate pGL4-basic constructs using Lipofectamine 2000 (Invitrogen). Luciferase activity was measured using the Beyotime RG027 Dual-Luciferase Reporter Gene Assay Kit.

Electrophoretic mobility shift assay

Electrophoretic mobility shift assay (EMSA) was performed using the LightShift Chemiluminescent EMSA Kit (#20148, Thermo Scientific) according to the manufacturer’s instructions. The sense probe sequences for EMSA are listed in Table S2. Biotinylated forward (F) and reverse (R) primers were diluted to 10 µM and probes were prepared according to the probe preparation system. The probes were then heated in a 95 °C water bath for 2 min, then allowed to cool naturally and anneal. The binding reaction mixture was incubated for 15 min at RT. Then, the biotinylated probe and recombinant protein were mixed and incubated at 15 °C for 25 min. DNA-protein complexes were separated by electrophoresis on a 6% non-denaturing polyacrylamide gel and transferred to a nylon membrane. The membrane was cross-linked using ultraviolet light, washed, and the signal was detected.

Evaluation of CD8 + T-cell-induced tumor cytotoxicity

CD8⁺ T cells were obtained from healthy donor PBMCs through a combination of Ficoll density-gradient centrifugation and magnetic bead isolation. These cells underwent a 14-day activation and expansion protocol in complete medium containing 10% FBS, anti-CD3/CD28 beads (Gibco), and 20 ng/mL recombinant human IL-2 (PeproTech). Subsequently, the expanded effector T cells were incubated with SPOCK1-overexpressing or vector-control tumor cells for 48 h at an effector-to-target (E:T) ratio of 3:1. The suspended CD8⁺ T cells were harvested after completion of the co-culture. The killing rate or cytotoxicity potential was evaluated by detecting the absorbance of the adherent target cells using the CCK-8 assay. Furthermore, co-culture supernatants were collected, and the secretion levels of granzyme B and IFN-γ were analyzed using specific ELISA kits.

Animal experiments

Animal experiments were conducted according to the protocols approved by the Ethics Committee of The Affiliated Kunshan Hospital of Jiangsu University. Five-to-six-week-old female BALB/c nude mice or female C57BL/6 J mice that were 5–6 weeks old were acquired from the Jiangsu University Animal Center and housed under SPF conditions at the Kunshan Affiliated Hospital Laboratory with standard feeding.

Subcutaneous xenograft studies in mice

Subcutaneous xenograft experiments in mice were conducted according to methods described in previous studies^29,30. In brief, female BALB/c nude mice or female C57BL/6 J mice (5 weeks old) received subcutaneous injections of 3 × 10⁶ pri-NSCLC or A549 cells transduced with either scrambled shRNA or sh-SPOCK1 in 100 µL of PBS. Additional mice were injected with pri-NSCLC or A549 cells expressing OE-SPOCK1 or a control vector (Vec) in a separate experiment. Tumor dimensions were recorded biweekly, and volumes were calculated using the formula (length × width²)/2.

Lung metastasis assay

In the tail vein metastasis assay, 3–5 × 10⁶ stable SPOCK1-knockdown or -overexpressing A549 cells (suspended in 100 μL sterile PBS) were injected into the tail veins of 5-week-old female BALB/c nude mice (n = 5). On day 30 post-inoculation, mice were euthanized. The intact lungs were immediately excised and fixed in 10% neutral buffered formalin (pH 7.4, Sigma-Aldrich) for 24 h. Standard hematoxylin and eosin (H&E) staining was performed (Harris hematoxylin for 5 min, 0.5% eosin ethanol counterstaining for 30 s) for histopathological validation. All sections were independently evaluated by two board-certified pathologists using a double-blind method to confirm metastatic features.

Statistical analysis

Statistical analysis was performed using GraphPad Prism (v9.5). Data were presented as mean ± standard deviation (SD). Significance of inter-group differences was evaluated via two-tailed Student’s t test or one-way ANOVA. Survival analysis was conducted using the Kaplan–Meier method. All experiments were conducted in triplicate. P < 0.05 was considered statistically significant.

Results

Expression and prognostic relevance of SPOCK1 in NSCLC

Although GWAS have identified multiple genetic loci for NSCLC susceptibility, significant differences in genetic backgrounds and environmental factors necessitate dedicated cohort studies in Chinese populations³¹. This will facilitate an in-depth investigation of pathogenic mechanisms for lung cancer based on gene-environment interactions and provide a scientific basis for precision medicine. Our hospital has successfully established the “Kunshan Elderly Population Cohort” biobank, containing 50,000 samples, in collaboration with a national key laboratory in Guangzhou. The genetic data for this cohort have been published previously^32–34. We performed a GWAS based on the Kunshan elderly population cohort to systematically elucidate the genetic susceptibility mechanisms of NSCLC development in Chinese populations and uncover disease-associated genetic variants and their potential molecular targets. The analysis included 380 NSCLC patients and 3800 healthy controls matched at a 1:10 ratio and aimed to clarify the heterogeneity of genetic risk factors for NSCLC. We identified one SNP with genome-wide significance (p < 5 × 10⁻⁸) and six SNPs with suggestive significance (p < 5 × 10⁻⁶) that were associated with the risk of NSCLC (Fig. 1A). A quantile-quantile (QQ) plot confirmed no substantial inflation due to population stratification (Fig. 1B). Figure 1C shows the detailed information of these SNPs. Annotation of these loci prioritized four candidate genes: SPOCK1 (rs4976398), THSD7B (rs72617049), CHRM5 (rs74007651), and EMC7 (rs80166484). To investigate the potential regulatory mechanisms linking these variants with gene function, we first searched the Genotype-Tissue Expression (GTEx) database and found that rs4976398 exhibited a significant cis-eQTL association with SPOCK1 expression in the normal lung tissue (P = 0.012; NES = 0.10); the risk allele (AG) was associated with lower SPOCK1 expression in the normal lung tissue (Supplementary Fig. 1A, B). We then validated this genotype-phenotype correlation in the clinical samples from the Kunshan elderly population cohort. Immunohistochemistry (IHC) results demonstrated that patients carrying the rs4976398 variant (AG) exhibited significantly higher SPOCK1 protein expression levels compared to non-carriers (AA), thereby confirming that the risk allele upregulates SPOCK1 in the tumors (Supplementary Fig. 1C). The discrepancy between the eQTL direction in normal tissue (GTEx) and the tumor patient cohort can be interpreted biologically. Gene regulation is highly context-dependent, and the regulatory landscape undergoes profound reprogramming during tumorigenesis³⁵. A systematic analysis by Gong et al., comparing TCGA and GTEx data demonstrated that a substantial proportion of cis-eQTLs are tumor-specific, which suggests that they are exclusively active or show different regulatory patterns in the tumor context compared to normal tissues³⁵. This supports our observation that the rs4976398-driven upregulation of SPOCK1 is potentially a tumor-specific event that is not captured by normal tissue databases.

Based on this combined bioinformatic and clinical evidence, we performed siRNA-mediated knockdown of the four candidate genes to assess their biological effects. Our results demonstrated that only SPOCK1 silencing significantly inhibited the viability of NSCLC cells (Supplementary Fig. 1D–G). This suggests a specific oncogenic role for SPOCK1 in NSCLC pathogenesis. Therefore, we selected SPOCK1 for further analysis.

TCGA database analysis demonstrated that SPOCK1 expression was significantly upregulated in lung adenocarcinoma (LUAD) (Supplementary Fig. 1H) and lung squamous cell carcinoma (Supplementary Fig. 1I) compared to normal lung tissues. Furthermore, high SPOCK1 expression levels positively correlated with the clinical stage and lymph node metastasis in NSCLC patients (Supplementary Fig. 1J–M). Moreover, validation analysis of the TCGA and GEO datasets (GSE30219 and GSE13213) demonstrated that elevated SPOCK1 expression levels were significantly associated with shorter overall survival (OS) in patients with NSCLC (Supplementary Fig. 1N–P). To eliminate the potential influence of confounding factors, we performed multivariate Cox proportional hazards analysis by adjusting for key clinical variables such as age, gender, and AJCC pathological stage. Multivariate analysis demonstrated that SPOCK1 expression remained significantly associated with OS even after adjusting for these clinical parameters (HR = 1.551, 95% CI: 1.290–2.115, p = 0.012). These results also demonstrated that SPOCK1 was an independent prognostic factor for patients with NSCLC (Supplementary Fig. 1Q).

To further validate the bioinformatics findings, we performed IHC analysis of tissue microarrays containing 90 paired cases of NSCLC tissues (T) and adjacent normal lung tissues (N) to evaluate the expression levels of SPOCK1 (Fig. 1D). Representative IHC images from two patients (Patients 1# and 2#) demonstrated significant overexpression of SPOCK1 in NSCLC tissues compared to adjacent non-cancerous tissues (p < 0.05, Fig. 1E, F). The overexpression of SPOCK1 was significantly related to adverse clinical outcomes (p < 0.05, Fig. 1G) and reduced survival rates in NSCLC patients (p < 0.05, Fig. 1H). Subsequently, we obtained fresh NSCLC tissues (T) and paired adjacent normal tissues (N) from a cohort of twelve locally-treated patients undergoing surgical intervention at regional hospitals. qRT-PCR (Fig. 1I) and western blotting (Fig. 1J) analysis of the tissue lysates of this cohort demonstrated significantly elevated levels of SPOCK1 mRNA and protein in the NSCLC tissues compared to the adjacent normal tissues. Further experimental analysis demonstrated higher SPOCK1 mRNA and protein levels in immortalized NSCLC cell lines (A549, H1650, H226, and H460) as well as primary NSCLC cells isolated from patients (priNSCLC) relative to human lung epithelial cells (HPAEpiC) (Fig. 1K, L). Immunofluorescence (IF) staining results using confocal laser scanning microscopy confirmed that SPOCK1 protein was predominantly localized in the cytoplasm of NSCLC cells (Fig. 1M). These findings suggest that SPOCK1 is a valuable biomarker for the diagnosis and treatment of NSCLC.

SPOCK1 silencing induces robust anti-NSCLC cell activity

We next investigated the functional role of SPOCK1 in NSCLC. We selected high SPOCK1-expressing priNSCLC and H460 cells for knockdown studies using lentiviral vectors encoding SPOCK1-specific shRNAs (sh-SPOCK1-s1 and sh-SPOCK1-s2), as well as a scrambled nonspecific control shRNA (shC). Knockdown efficacy was confirmed by significant reduction of SPOCK1 mRNA and protein levels in the SPOCK1-knockdown NSCLC cells (Fig. 2A, B). Transwell and Matrigel Transwell assays demonstrated that SPOCK1 knockdown suppressed NSCLC cell invasiveness and migratory capacity (Fig. 2C, D). CCK-8 and EdU assays demonstrated that SPOCK1 knockdown significantly reduced cell growth and proliferation (Fig. 2E, F). Furthermore, FACS-based cell cycle analysis with PI demonstrated that SPOCK1 knockdown increased the proportion of cells in the G1 phase and decreased the proportion of cells in the S phase, thereby indicating G1-S phase arrest (Fig. 2G). However, knockdown of SPOCK1 did not significantly alter the proliferation and migration of normal lung epithelial cells (Supplementary Fig. 2A, B). To further investigate the role of SPOCK1 on the in vivo growth of NSCLC cells, we established xenograft NSCLC tumor models by subcutaneously implanting priNSCLC expressing sh-SPOCK1-s1/s2 or shC into the flanks of nude mice. Tumor growth was assessed periodically, and tumor volumes were measured once every 7 days. Xenografts with sh-SPOCK1-s1/s2-transfected priNSCLC cells demonstrated significantly reduced tumor volumes compared to the shC group (Fig. 2H). After 28 days, the mice were euthanized and all xenograft tumors were excised and individually weighed. Tumors from the sh-SPOCK1-s1/s2 group were significantly lighter and smaller than those from the shC group (Fig. 2I, J). However, there were no significant differences in the body weights of mice belonging to the two groups (Fig. 2K). This suggests that the treatment did not adversely affect the overall health of the mice. Furthermore, SPOCK1 protein levels were significantly lower in the fresh tumors from the sh-SPOCK1 group compared to the shC group (Fig. 2L). IHC analysis of xenograft tissues demonstrated significantly lower Ki-67-positive staining in the sh-SPOCK1 group compared to the shC group (Fig. 2M). This suggested that SPOCK1 silencing suppressed in vivo NSCLC cell proliferation. To systematically evaluate the regulatory role of SPOCK1 in tumor metastasis, this study established a lung metastasis model via tail vein injection. Specifically, 5 × 10⁶ SPOCK1-shRNA-transfected A549 cells (suspended in 100 μL sterile PBS) were injected into the tail veins of nude mice. On day 30 post-inoculation, the mice were euthanized. The lungs were immediately harvested and fixed in 10% neutral buffered formalin. Standard H&E staining was performed for histopathological validation (Fig. 2N). All sections were independently evaluated by two board-certified pathologists using a double-blind method to confirm metastatic features. In vivo tail vein metastasis assay results demonstrated that SPOCK1 knockdown significantly reduced both lung metastasis incidence and metastatic nodule formation (Fig. 2O). In summary, these results demonstrate that SPOCK1 silencing significantly inhibits NSCLC malignant progression.

Fig. 2 — A, B Western blotting and qRT-PCR analysis show the expression levels of SPOCK1 protein and transcripts after lentivirus-mediated SPOCK1 knockdown in NSCLC cells. C, D Representative images show the results of Transwell migration and invasion abilities of control and SPOCK1-knockdown NSCLC cells. E, F Representative images of the CCK-8 and EdU assay results show the proliferation status of control and SPOCK1-knockdown NSCLC cells. G Flow cytometry results show the cell cycle distribution of the control and SPOCK1-knockdown NSCLC cells. Representative histograms show the percentage of cells in the G1 and S phases. H–K Subcutaneous tumor growth curves, tumor volume/weight, and body weight changes of nude mice in the control and SPOCK1 knockdown group. L Representative western blot (WB) images and quantitative analysis of SPOCK1 protein expression levels in subcutaneous tumor tissues from the control and SPOCK1-knockdown groups of mice. M Representative immunohistochemical (IHC) images show SPOCK1 and Ki67 expression levels in the xenograft tumor tissues from the control and SPOCK1-knockdown groups of mice. N Representative gross morphology of lung surface metastases and H&E-stained images of pulmonary tissue samples from the control and SPOCK1-knockdown groups of mice. O The number of pulmonary metastatic nodules in the H&E-stained sections from the control and SPOCK1-knockdown groups of mice. The cell experiments were repeated three times (n = 3); animal experiments included five mice per group (n = 5). Scale bar: 100 μm. shC control shRNA. *P < 0.05; N.S. not significant or P > 0.05.

Overexpression of SPOCK1 promotes tumorigenic activity in NSCLC cells

Since SPOCK1 silencing significantly inhibits NSCLC cell growth, we hypothesized that ectopic overexpression of SPOCK1 confers oncogenic activity. To test this hypothesis, priNSCLC, A549, and H1650 cells were infected with a lentiviral SPOCK1 cDNA-expression construct. qRT-PCR and western blotting results confirmed successful generation of stable SPOCK1-overexpressing cells, designated as OE-SPOCK1, compared to vector control (Vec) cells (Fig. 3A–C). Consistent with our hypothesis, EdU assays demonstrated that SPOCK1 overexpression significantly increased NSCLC cell proliferation (Fig. 3D–F). Subsequently, Transwell and Matrigel Transwell assay results demonstrated that OE-SPOCK1 cells exhibited accelerated migration and invasion in vitro (Fig. 3G–L). Furthermore, cell cycle analysis demonstrated that overexpression of SPOCK1 decreased the percentage of cells in the G1 phase and increased the percentage of cells in the S phase, thereby indicating accelerated G1-S cell cycle progression in the OE-SPOCK1 A549 cells compared to the control group (Fig. 3M). To evaluate the impact of SPOCK1 overexpression on NSCLC cell growth in vivo, priNSCLC cells transduced with the OE-SPOCK1 construct or Vector control were subcutaneously injected into the flanks of nude mice. Tumor growth was monitored periodically, and tumor volumes were measured every 7 days (Fig. 3N). On the 28th day of the experiment, all mice were euthanized, and the priNSCLC xenografts were excised and weighed. The tumor volumes and weights of OE-SPOCK1 priNSCLC xenograft tumors were significantly higher than the Vec priNSCLC xenografts (Fig. 3O, P). There were no significant differences in the body weights between the two groups (Fig. 3Q). Tumor tissue lysates from the OE-SPOCK1 group demonstrated significant upregulation of SPOCK1 protein levels compared to the Vec group (Fig. 3R). IHC staining demonstrated higher Ki-67 expression in the OE-SPOCK1 group, thereby indicating enhanced tumor cell proliferation in vivo (Fig. 3S). To elucidate the functional role of SPOCK1 in tumor metastasis, a lung metastasis model was generated by tail vein injection of SPOCK1-overexpressing A549 cells (SPOCK1-OE group) or empty vector-transfected control cells (Vec group) into immunodeficient nude mice. SPOCK1 overexpression enhanced metastatic dissemination, leading to a higher lung metastasis burden (Fig. 3T, U). Collectively, both in vitro and in vivo genetic gain-of-function studies demonstrated that SPOCK1 was a critical regulator of NSCLC tumorigenicity, with its overexpression sufficient to confer malignant progression. These findings further support the oncogenic role of SPOCK1 in NSCLC pathogenesis and highlight its potential as a therapeutic target.

Fig. 3 — A–C Western blotting and qRT-PCR analysis of the SPOCK1 protein and mRNA expression levels in control and SPOCK1-overexpressing NSCLC cells. D–F Representative images show EdU assay results for the control and SPOCK1-overexpressing NSCLC cells. G–L Representative images show the results of Transwell migration and invasion assays for the control and SPOCK1-overexpressing NSCLC cells. M Representative cell cycle assay results for the control and SPOCK1-overexpressing NSCLC cells. N–Q Subcutaneous tumor growth curves, tumor weights, and body weight changes in nude mice xenografted with control and SPOCK1-overexpressing NSCLC cells. R SPOCK1 protein levels in tumor tissue lysates from the nude mice xenografted with control and SPOCK1-overexpressing NSCLC cells. S Immunohistochemical staining (IHC) results show SPOCK1 and Ki-67 protein expression levels in tumor nodules from the nude mice xenografted with control and SPOCK1-overexpressing NSCLC cells. T Representative gross morphology of lung surface metastases and H&E-stained images of pulmonary tissue samples from the nude mice xenografted with control and SPOCK1-overexpressing NSCLC cells. U Number of pulmonary metastatic nodules in H&E-stained sections from nude mice xenografted with control and SPOCK1-overexpressing NSCLC cells. Cell experiments were performed in triplicate (n = 3); animal experiments were performed with five mice per group (n = 5). Scale bar: 100 μm. Abbreviations: Vec, empty vector. *P < 0.05; N.S. not significant or P > 0.05.

Proteomic and metabolomic analyses link SPOCK1 to ferroptosis regulation in NSCLC

The progression of NSCLC is critically dependent on metabolic reprogramming, wherein glucose, lipid, and amino acid metabolism are remodeled in the tumor cells to synergistically meet their bioenergetic demands, supply biosynthetic precursors, and maintain redox homeostasis^36–40. SPOCK1 is a critical regulator of glucose and lipid metabolism^12,41,42. SPOCK1 enhances tumor cell proliferation by enhancing glucose uptake and lactate production, thereby suggesting its role in coordinating metabolic remodeling to drive malignant progression¹². Moreover, RBM15-mediated m6A modification stabilizes SPOCK1 transcripts, consequently facilitating malignant progression¹³. Despite SPOCK1’s established role in orchestrating the metabolic pathways that govern cellular glucose and lipid metabolism, its link to ferroptosis is not known. We hypothesized that SPOCK1 drives NSCLC progression by orchestrating metabolic reprogramming, a process contingent on dynamic changes in the protein expression profile. To systematically investigate the metabolic-protein regulatory network mediated by SPOCK1, we conducted integrated proteomic and metabolomic profiling of SPOCK1-silenced A549 cells.

Metabolic reprogramming in cancer is frequently driven by the activity of metabolic enzymes. Therefore, we focused on identifying proteomic and metabolomic alterations associated with SPOCK1 knockdown. Proteomic analysis demonstrated 1181 differentially expressed proteins between sh-SPOCK1 and shC groups, as shown in the circular heatmap (Fig. 4A). The Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis demonstrated that the ferroptosis signaling pathway was significantly enriched among the differentially expressed genes in the SPOCK1 knockdown group compared to the control group (Fig. 4B). Furthermore, Gene Set Enrichment Analysis (GSEA) results showed that the ferroptosis pathway was significantly enriched in the SPOCK1 knockdown group (Fig. 4C). The volcano plot and hierarchical clustering analysis results showed that the expression levels of multiple ferroptosis-related genes in the GSH synthesis pathway were significantly downregulated in the SPOCK1 knockdown cells (Fig. 4D, E). Subsequently, metabolomic profiling was performed to complement proteomic findings. PCA demonstrated clear separation between sh-SPOCK1 and shC groups, indicating global metabolomic remodeling (Fig. 4F). Hierarchical clustering demonstrated 144 downregulated and 87 upregulated metabolites in the sh-SPOCK1 group compared with the shC group (Fig. 4G, H). Volcano plots showed distinct metabolite signatures for the sh-SPOCK1 group (Fig. 4I). Integrated pathway analysis of proteomic and metabolomic data and annotation with the KEGG pathway resulted in the identification of 56 overlapping pathways, with ferroptosis emerging as one of the significantly altered signaling pathways (Fig. 4J, K). The enriched and differentially expressed metabolites and proteins in the ferroptosis signaling pathway are shown in Fig. 4L. Proteomic and metabolomic analysis demonstrated significant alterations in the GSH metabolic network, specifically characterized by significant differential expression of multiple key regulatory proteins in the GSH metabolic pathway (Fig. 4L). These include the cystine/glutamate antiporter subunit SLC3A2, glutamate-cysteine ligase catalytic subunit (GCLC) and its regulatory subunit (GCLM), as well as GSH peroxidase 4 (GPX4). These findings suggest that SPOCK1 plays a central role in regulating ferroptosis and GSH metabolism.

Fig. 4 — A Circular heatmap of differentially expressed genes (DEGs) between sh-SPOCK1 and shC groups (n = 3). B Bubble plot shows enriched KEGG pathways for the DEGs. As shown, ferroptosis is a key significantly enriched pathway. C GSEA plot shows enrichment of the ferroptosis pathway in the sh-SPOCK1 group compared to the shC group. D, E Volcano plot and hierarchical clustering of ferroptosis-related genes. F Principal component analysis (PCA) plot shows clear separation between sh-SPOCK1 and shC groups (n = 7). G–I Heatmap and volcano plots of differentially expressed metabolites. J, K Venn diagrams and pathway lists integrating protein and metabolite changes show the pathways co-regulated at both levels. L Differential expressed proteins and metabolites implicated in the ferroptosis pathway in the sh-SPOCK1 and shC groups. *P < 0.05; N.S, not significant or P > 0.05.

SPOCK1 silencing promotes ferroptosis in NSCLC cells

To determine whether SPOCK1 regulates ferroptosis in NSCLC cells, we first assessed cell viability in SPOCK1-silenced cells treated with ferroptosis-specific inhibitor ferrostatin-1 (Fer-1) as well as inhibitors of apoptosis (ZVAD-FMK), necroptosis (necrostatin-1), and autophagy (3-MA). Fer-1 restored viability of the SPOCK1-silenced cells, whereas other inhibitors did not show significant effects (Supplementary Fig. 2C). This confirmed that ferroptosis was the predominant cell death mechanism in the SPOCK1 knockdown cells. Ferroptosis is characterized by iron overload and LPO. We used FerroOrange and C11-BODIPY probes and detected increased intracellular Fe²⁺ (Fig. 5A, B) and LPO levels (Fig. 5C, D) in the SPOCK1-silenced cells compared to the shC group, but these effects were attenuated by Fer-1 treatment, as demonstrated by IF staining (Supplementary Fig. 2D, E). Ultrastructural analysis demonstrated mitochondrial fragmentation and cristae enlargement—characteristic morphological hallmarks of ferroptosis activation. These alterations were accompanied by increased ROS production, which directly triggers LPO accumulation and amplified ferroptotic cell death⁴³. Transmission electron microscopy (TEM) was used to determine mitochondrial ultrastructural changes and investigate whether SPOCK1 silencing drives ferroptosis via mitochondrial dysfunction. TEM imaging demonstrated that SPOCK1 knockdown induced mitochondrial atrophy, which was characterized by the loss of mitochondrial cristae and dissolution of bilayer membrane integrity (as indicated by the yellow arrow) (Fig. 5E, F). Furthermore, mitochondrial superoxide levels were significantly elevated in the SPOCK1-silenced cells (Fig. 5G, H).

Fig. 5 — A, B Intracellular Fe²⁺ levels in the sh-SPOCK1 and shC groups of NSCLC cells. Scale bar: 20 µm (n = 3). C, D Representative C11-BODIPY immunofluorescence images show lipid peroxidation in the sh-SPOCK1 and shC groups of NSCLC cells. Scale bar: 20 µm (n = 3). E, F Transmission electron microscopy (TEM) images show ultrastructural details of the mitochondria in the sh-SPOCK1 and shC groups of NSCLC cells. As shown, sh-SPOCK1-transfected NSCLC cells show disruption of mitochondrial membrane and loss of cristae loss (indicated by the yellow arrow). N cell nucleus, M mitochondria, ER endoplasmic reticulum. Scale bar: 500 nm. G, H Mito-SOX (red) and Hoechst (blue) immunofluorescence staining shows mitochondrial superoxide levels in the sh-SPOCK1 and shC groups of NSCLC cells. Scale bar: 20 µm (n = 3). I, J Estimation of glutathione (GSH) and oxidized glutathione (GSSG) levels in the sh-SPOCK1 and shC groups of NSCLC cells (n = 3). K, L DCFH-DA staining results show intracellular reactive oxygen species (ROS) levels in the sh-SPOCK1 and shC groups of NSCLC cells (n = 3). M, N MDA assay results show malondialdehyde (MDA) levels in the sh-SPOCK1 and shC groups of NSCLC cells (n = 3). *P < 0.05; NS not significant or P > 0.05.

GSH is a key antioxidant in the GSH/GSSG redox cycle and safeguards cancer cells against oxidative damage by neutralizing ROS. Depletion of GSH and concomitant accumulation of MDA, a terminal LPO byproduct, are established biomarkers of ferroptosis⁴⁴. To assess SPOCK1’s role in GSH metabolism, we quantified intracellular GSH and oxidized GSSG levels. SPOCK1 silencing significantly reduced total GSH levels and increased the levels of GSSG (Fig. 5I, J), thereby indicating impaired GSH synthesis and redox imbalance. Consistent with this, SPOCK1-silenced cells showed higher ROS levels, as assessed by DCFH-DA staining (Fig. 5K, L) and MDA accumulation (Fig. 5M, N) compared to the control cells.

Collectively, these data demonstrate that SPOCK1 silencing activates the ferroptosis pathway in NSCLC cells through coordinated disruption of GSH metabolism. This mechanistic link highlights the significant role of SPOCK1 as a potential regulator of ferroptosis and offers novel insights into its role in NSCLC progression and therapeutic targeting.

SPOCK1 regulates GSH biosynthesis by upregulating SLC3A2

To define the mechanism by which SPOCK1 inhibition induces ferroptosis in NSCLC cells, we focused on its role in regulating GSH biosynthesis. The protein expression levels of key molecules in GSH metabolism (SLC7A11, SLC3A2, GPX4, GCLC, and GCLM) were evaluated in the control and SPOCK1 knockdown cells using western blotting. Consistent with the proteomic analysis, SPOCK1 knockdown cells showed significantly lower expression levels of SLC3A2, GPX4, GCLC, and GCLM proteins (Fig. 6A). SLC3A2 is a key initiator of GSH synthesis, but its expression was significantly suppressed in the SPOCK1 knockdown cells. Therefore, we selected SLC3A2 as a downstream target gene of SPOCK1 for subsequent research. To validate whether SPOCK1 regulates the expression of SLC3A2, we analyzed the mRNA and protein levels of SLC3A2 in the control and OE-SPOCK1 cells. SPOCK1 overexpression significantly increased the levels of SLC3A2 transcripts and protein without altering SLC7A11 levels (Fig. 6B, C). This suggested that SPOCK1 selectively regulates SLC3A2 at the transcriptional level. To establish functional causality, we next determined if SLC3A2 was required for the anti-ferroptotic function of SPOCK1. Exogenous overexpression of SLC3A2 (sh-SPOCK1-s1 + SLC3A2) rescued the decreased protein expression levels of GPX4, GCLC, and GCLM in the SPOCK1 knockdown cells without affecting SPOCK1 expression itself (Fig. 6D). Therefore, SPOCK1 likely maintains GSH biosynthesis by regulating SLC3A2. Functionally, overexpression of SLC3A2 significantly reverses cell cycle arrest (Fig. 6E) and restores redox homeostasis by normalizing GSH levels and decreasing MDA accumulation and LPO in SPOCK1-silenced cells (Fig. 6F–H). These results suggest that SPOCK1 transcriptionally upregulates SLC3A2 to sustain GSH biosynthesis and is therefore associated with the ferroptosis-resistant phenotype in NSCLC.

Fig. 6 — A Western blot analysis of SLC7A11, SLC3A2, GCLC, GCLM, and GPX4 protein expression in the control and SPOCK1-knockdown priNSCLC and A549 cells. B SLC7A11 and SLC3A2 protein levels in the control and SPOCK1-overexpressing cells. C qPCR analysis of relative SLC3A2 mRNA expression in the control and SPOCK1-knockdown or SPOCK1-overexpressing NSCLC cells. D The levels of SLC3A2, GCLC, GCLM, and GPX4 proteins in the SPOCK1 knockdown and SLC3A2-overexpressing SPOCK1 knockdown cells. E Representative images show the cell proliferation analysis of different groups of NSCLC cells. Scale bar: 100 µm. F Malondialdehyde (MDA) and G lipid peroxidation (BODIPY™ 581/591 C11 staining) levels in various treatment groups. Scale bar: 20 µm. H Glutathione (GSH) and oxidized glutathione (GSSG) levels in all the different groups of NSCLC cells. All experiments were conducted with triplicate samples per group. ROS reactive oxygen species. *P < 0.05; N.S. not significant or P > 0.05.

SPOCK1 upregulates SLC3A2 expression by activating the JAK2/STAT3 pathway

Our findings indicate that SPOCK1 inhibits ferroptosis in NSCLC cells by transcriptionally upregulating SLC3A2 expression. However, since SPOCK1 itself is not a transcription factor, it cannot directly initiate the transcription of SLC3A2. Therefore, we hypothesize that SPOCK1 activates one or more specific transcription factors, thereby indirectly regulating the transcriptional level of SLC3A2. Proteomic sequencing analysis demonstrated that significantly downregulated genes in the SPOCK1-knockdown cells were closely associated with the JAK-STAT signaling pathway according to KEGG pathway enrichment analysis (Fig. 7A). Since JAK2/STAT3 signaling plays a key role in the progression of NSCLC^45–47, we hypothesized that SPOCK1 regulates JAK2-STAT3 signaling pathway activity. Western blot results demonstrated that SPOCK1 silencing significantly decreased phosphorylation of JAK2 (Y1007/Y1008) and STAT3 (Tyr705) in the NSCLC cells without altering the total protein levels of JAK2 and STAT3 (Fig. 7B), whereas SPOCK1 overexpression enhanced phosphorylation of JAK2 and STAT3 (Fig. 7C). This demonstrates that SPOCK1 activates the JAK2/STAT3 pathway in NSCLC cells. SPOCK1 is a component of the extracellular matrix (ECM) and belongs to the SPARC family of matricellular proteins, which function primarily by modulating ECM-integrin interactions^48,49. Integrins are the principal transmembrane bridges connecting the extracellular environment to intracellular signaling cascades, representing the most important canonical pathway through which ECM proteins exert their effects^50,51. A recent study by Du et al. reported that SPOCK1 promotes activation and proliferation of hepatic stellate cells by interacting with integrin α5β1 to activate the PI3K/Akt pathway⁵². JAK2 is canonically activated by cytokine receptors (e.g., IL-6R) as well as a non-canonical activation pathway wherein integrin clustering triggers Focal Adhesion Kinase (FAK), which subsequently recruits and activates Src kinase^53,54. Notably, Src kinase is a known direct activator of JAK2 capable of inducing its phosphorylation^54–56. Since integrin α5β1 acts as a canonical upstream activator of the FAK/Src complex and SPOCK1 serves as a ligand for this integrin, we hypothesize that SPOCK1 bind to integrin α5β1 and initiates a signaling cascade that activates the FAK/Src complex, which in turn triggers the JAK2/STAT3 pathway via non-canonical mechanisms.

Fig. 7 — A Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis of the downregulated differentially expressed genes (DEGs) in the sh-SPOCK1 cells compared to the shC cells. B, C Western blot (WB) analysis of the JAK2/STAT3 pathway activation in the control and SPOCK1-knockdown or SPOCK1-overexpressing NSCLC cells. D, E Representative images show immunofluorescence staining of STAT3 activation in the control and SPOCK1-knockdown or SPOCK1-overexpressing NSCLC cells. Scale bar: 20 µm. F Representative western blot (WB) images show the nuclear and cytoplasmic distribution of p-STAT3 in the control and SPOCK1-knockdown or SPOCK1-overexpressing NSCLC cells. G–J SLC3A2 mRNA and protein expression levels in NSCLC cells treated with Stattic (STAT3 inhibitor, 0.5–4 µM). K Chromatin immunoprecipitation (ChIP) assay shows p-STAT3 binding to the SLC3A2 promoter region. L The JASPAR database analysis shows the predicted highest-scoring binding site of STAT3 on the SLC3A2 promoter. M, N Schematic representation of STAT3 binding sites in the wild-type (WT) and mutant (MUT) SLC3A2 promoters used in dual-luciferase assays. All experiments were performed in triplicate per group. *P < 0.05; N.S. not significant or P > 0.05.

To validate this hypothesis, we first examined the physical interaction between SPOCK1 and integrin α5β1. Co-IP assays confirmed direct interaction between SPOCK1 and integrin α5β1 (Supplementary Fig. 2F), a finding further corroborated by immunofluorescence (IF) analysis showing distinct spatial colocalization of SPOCK1 and integrin β1 (Supplementary Fig. 2G). We next investigated whether this interaction was functionally required for downstream signaling. SPOCK1 overexpression significantly increased the phosphorylation of FAK (Tyr397) and Src (Tyr416), indicating activation of the proximal integrin signaling complex and subsequent activation of the JAK2/STAT3 signaling pathway (Supplementary Fig. 2H). To definitively establish causality, we treated SPOCK1-overexpressing cells with specific neutralizing antibodies against integrin α5β1 (anti-α5) and (anti-β1). Western blotting results demonstrated that blocking either subunit significantly abrogated SPOCK1-induced phosphorylation of FAK and Src, and subsequent activation of the JAK2/STAT3 pathway (Supplementary Fig. 2H). The combined administration of anti-α5 and anti-β1 exerted a synergistic inhibitory effect. Collectively, these data provide robust evidence that SPOCK1 activates the JAK2/STAT3 signaling pathway by binding to cell-surface integrin α5β1 and triggering the intracellular FAK/Src cascade.

Since phosphorylation of STAT3 (Tyr705) enhances its nuclear translocation and activates its role as a transcription factor⁵⁷, we investigated the effect of SPOCK1 expression on the subcellular localization of STAT3 in A549 cells using immunofluorescence assays. STAT3 levels in the nucleus were significantly lower in the SPOCK1 knockdown cells, whereas overexpression of SPOCK1 significantly increased the levels of STAT3 in the nucleus (Fig. 7D, E). These findings further confirm that SPOCK1 facilitates nuclear translocation of STAT3, thereby activating the STAT3 signaling pathway. Furthermore, western blotting results showed that SPOCK1 gene silencing significantly inhibited the translocation of phosphorylated STAT3 (p-STAT3) from the cytoplasm to the nucleus. Conversely, overexpression of SPOCK1 significantly increased nuclear translocation of p-STAT3 (Fig. 7F). These findings suggest that SPOCK1 drives the nuclear translocation of STAT3, thereby enhancing its transcription factor function. Since STAT3 regulates the expression of ferroptosis-related genes such as GPX4 and SLC7A11^58,59, we hypothesized that STAT3 directly regulates SLC3A2 mRNA expression level. STAT3 inhibition by Stattic, a potent STAT3 inhibitor that suppresses STAT3 phosphorylation, significantly reduced SLC3A2 mRNA and protein levels in the NSCLC cells (Fig.7G, H). This indicated a correlation between SLC3A2 and STAT3. Furthermore, SPOCK1-driven SLC3A2 upregulation was abolished by Stattic, thereby confirming STAT3 as the intermediary axis (Fig. 7I, J). To further confirm whether STAT3 directly regulates SLC3A2 expression, potential STAT3 binding sites within the SLC3A2 promoter region were predicted using JASPAR (Table S2). Several putative STAT3 binding sites were identified, as shown in Table S2. Chromatin immunoprecipitation (ChIP) assays confirmed direct interaction between p-STAT3 and SLC3A2 (Fig. 7K). Subsequently, to validate the specificity of STAT3 binding at the predicted site, we introduced mutations into the site with the highest JASPAR prediction score (Fig. 7L, M) and conducted dual-luciferase reporter gene assays. The results showed that when STAT3 was overexpressed, luciferase activity driven by the WT promoter was significantly higher than that of the control group, but a significant increase in luciferase activity was not observed when this site was mutated (Fig. 7N). These results demonstrate that STAT3 specifically binds to the SLC3A2 promoter and activates its transcription. Collectively, this study confirms that the SPOCK1/JAK2/STAT3 axis mediates transcriptional upregulation of SLC3A2 in NSCLC, thereby highlighting its critical role in ferroptosis resistance.

SPOCK1 silencing enhances erastin-induced ferroptosis synergistically

Erastin, a known ferroptosis inducer, inhibits the cysteine/glutamate antiporter and disrupts GSH synthesis⁶⁰. The effects of SPOCK1 on SLC3A2 expression suggest that SPOCK1 expression levels modulate Erastin sensitivity in the NSCLC cells. To determine this, we stably knocked down SPOCK1 in the A549 cells using a lentiviral vector and treated them with Erastin. EdU and Transwell assays showed that Erastin treatment of SPOCK1 knockdown cells significantly reduced their proliferation and migration abilities (Fig. 8A, B). To determine whether SPOCK1 suppression enhances the anti-cancer activity of Erastin in vivo, SPOCK1 knockdown A549 cells were subcutaneously implanted into nude mice. Then the mice were treated with Erastin using intraperitoneal injections (15 mg/kg/every 3 days). On day 28, the mice were euthanized, and the tumors were excised and weighed. Tumor volumes and weights were significantly reduced in the SPOCK1-depleted group treated with Erastin compared to the controls, but the body weights of the two groups were comparable (Fig. 8C–F). Furthermore, IHC staining showed significantly higher levels of 4-hydroxynonenal (4HNE), a marker of LPO, in both the control and SPOCK1-depleted cells treated with Erastin, but 4HNE staining was stronger in SPOCK1-depleted group compared to the control group (Fig. 8G). This further confirmed that SPOCK1 silencing enhanced Erastin-induced ferroptosis in vivo. Furthermore, IHC staining results showed significantly lower Ki-67 and SLC3A2 expression in the tumors derived from the Erastin-treated SPOCK1 knockdown group. Western blotting (WB) results also demonstrated that the protein levels of 4HNE were significantly higher and the levels of Ki-67, SLC3A2, and GPX4 were significantly lower in xenograft tissues derived from the Erastin-treated SPOCK1 knockdown group (Fig. 8H). To investigate the synergistic anti-metastatic effect of SPOCK1 knockdown combined with Erastin treatment, a tail vein metastasis model was established by injecting SPOCK1-knockdown A549 cells into nude mice. Erastin (15 mg/kg) or vehicle control was administered via intraperitoneal injection every three days for 30 days. Histopathological analysis of lung tissues demonstrated that the formation of metastatic foci was significantly lower in the Erastin-treated SPOCK1 knockdown group compared to either intervention alone, thereby demonstrating cooperative therapeutic potential in NSCLC metastasis (Fig. 8I, J).

Fig. 8 — A, B Cell proliferation and migration assay results for sh-SPOCK1-s1-transfected A549 cells treated with 10 µM Erastin or 0.1% DMSO as a vehicle control. C–F Tumor volumes, tumor weights, and body weights of mice xenografted with SPOCK1-knockdown A549 cells treated with Erastin or vehicle. G Representative IHC staining images show the expression levels of 4HNE, SLC3A2 and Ki67 in tumor tissues from mice xenografted with SPOCK1-knockdown A549 cells treated with Erastin or vehicle. H Representative images of western blot analysis show the expression levels of 4HNE, SLC3A2, GPX4, and Ki67 proteins in the xenograft tumor tissues from mice xenografted with SPOCK1-knockdown A549 cells treated with Erastin or vehicle. I The gross morphology of pulmonary surface metastases and H&E-stained images of lung tissue samples from the mice xenografted with SPOCK1-knockdown A549 cells treated with Erastin or vehicle. J The number of lung metastatic nodules in H&E-stained sections of lung tissue samples from the mice xenografted with SPOCK1-knockdown A549 cells treated with Erastin or vehicle. Scale bar: 100 µm. *P < 0.05; N.S. not significant or P > 0.05.

Collectively, these results conclusively demonstrate that the combination of SPOCK1 silencing and treatment with the ferroptosis inducer Erastin significantly suppresses growth and metastatic progression of NSCLC.

HSF1 is a master transcription factor of SPOCK1 in NSCLC cells

In NSCLC, SPOCK1 is significantly upregulated at both the transcriptional (mRNA) and translational (protein) levels. This abnormal overexpression pattern is closely associated with the malignant biological behaviors of tumor cells, including enhanced cell proliferation, invasion, and metastatic capabilities, and is significantly correlated with poor clinical prognosis. To elucidate the regulatory mechanisms underlying SPOCK1 dysregulation in NSCLC, we conducted an integrated bioinformatics analysis using the Transcription Factor Binding Database (TFBIND), Gene Transcription Regulation Database (GTRD), and Prediction of Transcription Factor Binding Sites (PROMO) databases. After comprehensive screening of transcription factor binding motifs and promoter regulatory elements, HSF1 (Heat Shock Factor 1) was identified as a putative master transcription factor regulating SPOCK1 expression (Fig. 9A). Gene Expression Profiling Interactive Analysis (GEPIA) demonstrated a positive correlation between HSF1 and SPOCK1 expression (Fig. 9B). Furthermore, survival analysis demonstrated that high HSF1 expression was associated with poor prognosis in NSCLC (Fig. 9C). This study systematically elucidated the unidirectional regulatory mechanism of HSF1 on SPOCK1 through functional experiments. HSF1 knockdown significantly reduced both SPOCK1 mRNA and protein levels (Fig. 9D, E), whereas HSF1 overexpression upregulated SPOCK1 mRNA and protein levels (Fig. 9F, G). However, SPOCK1 knockdown or overexpression did not significantly alter HSF1 expression levels, thereby confirming the unidirectional regulation of SPOCK1 by HSF1 (Fig. 9H). Since the regulatory effects of HSF1 on SPOCK1 were consistent at both mRNA and protein levels, it is plausible that HSF1 directly regulates SPOCK1 promoter activity rather than acting through post-transcriptional mechanisms. We identified five putative HSF1-binding sites (HBSs) in the SPOCK1 promoter region using JASPAR (Fig. 9I, Table S4). To validate these findings, SPOCK1-WT and SPOCK1-MUT promoter luciferase reporter vectors were constructed by mutating the predicted HBSs (HBSa/b/c) within the SPOCK1 promoter sequence (bases 1317–1550) (Fig. 9J, K). HSF1 overexpression significantly increased luciferase activity of the SPOCK1-WT vector but did not significantly alter luciferase activity of the SPOCK1-MUT vector (Fig. 9K). EMSA with synthesized oligonucleotides covering the presumed binding (WT probe) and mutated sites (MUT probe) confirmed specific HSF1 binding to the SPOCK1 promoter at HBSa/c (Fig. 9L). Collectively, these findings demonstrate that HSF1 acts as a transcription factor of SPOCK1 in NSCLC.

Fig. 9 — A TFBIND, GTRD, and PROMO database analysis results show the potential transcription factors binding to the SPOCK1 promoter. B Gene Expression Profiling Interactive Analysis (GEPIA) results show the correlation between HSF1 and SPOCK1 in NSCLC. C Kaplan–Meier survival analysis results show the overall survival rates of NSCLC patients from the TCGA dataset based on high and low HSF1 expression. D qRT-PCR analysis of the expression levels of HSF1 and SPOCK1 transcripts in control and HSF1 knockdown NSCLC cells. E Representative western blot and quantitative analysis of HSF1 and SPOCK1 protein expression levels in control and HSF1 knockdown NSCLC cells. F Relative HSF1 and SPOCK1 mRNA expression levels in control and HSF1-overexpressing NSCLC cells. G SPOCK1and HSF1 protein expression levels in control and HSF1-overexpressing NSCLC cells. H HSF1 protein expression levels in control and SPOCK1 knockdown NSCLC cells. I, J JASPAR database analysis shows the putative HSF1 binding site in the SPOCK1 promoter. K Luciferase reporter assay results for the interaction of HSF1 with wild-type (WT-SPOCK1) and mutated SPOCK1 (MUT-SPOCK1) promoter sequences. L EMSA results validate that HSF1 binds to the SPOCK1 promoter. Data are presented as means ± SDs of three independent experiments (n = 3). *P < 0.05; N.S. not significant or P > 0.05.

To investigate whether HSF1 drives SPOCK1 expression and subsequent inhibition of ferroptosis, we performed rescue assays by transfecting HSF1-overexpressing cells with SPOCK1 shRNA. WB analysis demonstrated that HSF1 overexpression significantly upregulated the expression levels of key anti-ferroptotic proteins, GPX4 and SLC3A2, but this upregulation was effectively abolished by SPOCK1 knockdown (Supplementary Fig. 3A). Functional assays demonstrated that HSF1 overexpression significantly decreased MDA levels, reduced intracellular Fe²⁺ accumulation, and attenuated lipid peroxidation, all of which indicated suppression of ferroptosis (Supplementary Fig. 3B–D). However, concurrent silencing of SPOCK1 significantly reversed the suppressive effects of HSF1 overexpression, leading to elevated levels of MDA, intracellular Fe²⁺, and lipid peroxidation, thereby restoring cellular sensitivity to ferroptosis (Supplementary Fig. 3B–D). Collectively, these data provide robust evidence that the HSF1-SPOCK1 axis inhibits ferroptosis in the NSCLC cells.

The expression of SPOCK1 in the NSCLC microenvironment is associated with the extent of immune cell infiltration

The immune microenvironment in NSCLC involves a dynamic interplay between immunosuppression and anti-tumor immunity, which ultimately determines disease progression and treatment response characteristics^61–63. Since SPOCK1 promotes tumor aggression by inhibiting ferroptosis, and TME remodeling is closely coupled with ferroptosis, we next investigated whether these intrinsic cellular changes influence the extrinsic tumor microenvironment (TME) and affect immune cell infiltration^64,65.

To test this hypothesis, we initially conducted a bioinformatics analysis of the TCGA-NSCLC cohort and identified a significant correlation between SPOCK1 expression and infiltration levels of specific immune cell subsets (Fig. 10A, B). In NSCLC tissues, the expression level of SPOCK1 correlated significantly with the infiltration levels of T-cell subsets such as Th17 and CD8⁺ T cells (Fig. 10A, B). This suggests that SPOCK1 participates in the immunoregulatory processes of the TME by modulating T cell-mediated immune responses. Next, we performed scRNA-seq of paired samples (tumor tissue and adjacent normal tissue) from seven NSCLC patients. After stringent quality control, we obtained 181,281 high-quality cells. Based on the expression profiles of classic marker genes, these cells were classified into 14 distinct types, including epithelial cells, endothelial cells, T cells, B cells, and smooth muscle cells (Fig. 10C). This study specifically focused on the potential regulatory role of SPOCK1 in T cell function. First, high-resolution clustering analysis of T cell subsets identified multiple functional subpopulations, including CD8⁺ T cells, CD4⁺ T cells, follicular helper T cells (Tfh), and Th17 cells (Fig. 10D). Subsequently, epithelial cells were divided into SPOCK1-positive and SPOCK1-negative groups based on their SPOCK1 expression levels. Based on the cell-cell interaction analysis using CellPhoneDB, SPOCK1-positive epithelial cells exhibited significant interactions with various T cell subsets in the tumor tissues compared to the normal tissues (Fig. 10E, F). This suggests that SPOCK1-positive epithelial cells (SPOCK1⁺ ECs) may contribute to shaping the immune microenvironment by regulating T cell function. Furthermore, differential gene enrichment analysis between the SPOCK1-positive and SPOCK1-negative groups of epithelial cells demonstrated significant enrichment of genes involved in GSH metabolism, ferroptosis, and the JAK/STAT signaling pathway, which aligns with our previous experimental results (Fig. 10G). Furthermore, PD-L1 and PD-1 checkpoint pathway in cancer also showed significant enrichment, thereby suggesting the involvement of SPOCK1 in immune regulatory functions (Fig. 10G).

Fig. 10 — A, B SPOCK1 expression correlates with multiple immune cell populations, as demonstrated by analysis of the TCGA-NSCLC cohort. C UMAP (Uniform Manifold Approximation and Projection) analysis of 181,281 cells from paired NSCLC samples (seven pairs of cancer tissues and adjacent normal tissues). D Identification of T-Cell subpopulations based on single-cell transcriptomic data. E, F Network map of cell-cell interactions between epithelial cells and various T cell subsets in paired NSCLC and adjacent normal lung tissues. G KEGG pathway analysis on differentially expressed genes (DEGs) between SPOCK1-positive cells and SPOCK1-negative cells. H, I Multiplex immunohistochemistry (mIHC) staining analysis of a tissue microarray (TMA) from NSCLC patients, presented according to SPOCK1 expression levels (high expression group (H) and low expression group (I). The Fig. includes a color-coded legend corresponding to the following markers: SPOCK1 (green), CD3 (cyan), CD4 (yellow), CD8 (orange), IL-17 (red), and p-STAT3 (magenta). Nuclei were counterstained with DAPI (Blue). Quantitative analysis after staining was used to compare the proportions of immune cell populations between the high- and low-SPOCK1 expression groups. These Figures show the percentages of CD3-positive cells (J), CD4-positive cells (K), CD8-positive cells (L), IL-17-positive cells (M) and p-STAT3-positive cells (N). The terms “Low-expression group” and “High-expression group” indicate the SPOCK1 low-expression and high-expression groups, respectively. The high-expression group included 71 samples and the low-expression group included 39 samples. * indicates P < 0.05 compared to the “Low-expression group.” The scale bar represents 200 μm. SPOCK1⁺ECs SPOCK1-positive epithelial cells, SPOCK1-ECs SPOCK1-negative epithelial cells, high exp high expression, low exp low expression.

To validate the above analysis and also investigate the potential interactions between SPOCK1 expression and immune cell infiltration, we performed mIHC to detect the expression levels of six marker proteins (SPOCK1, CD3, CD4, CD8, P-STAT3, and IL-17) in NSCLC tissues. Based on SPOCK1 expression levels, the tissue samples were categorized into a high-expression group (SPOCK1-high) and a low-expression group (SPOCK1-low) (Fig. 10H, I). The mIHC analysis results showed that the SPOCK1-high group exhibited significantly higher infiltration of the CD4⁺ T and CD8⁺ T cells than the SPOCK1-low group, whereas infiltration levels of CD3⁺ T cells did not show significant changes (Fig. 10J–L). Furthermore, the secretion of IL-17 by the Th17 cells was significantly decreased in the SPOCK1 high-expression group (Fig. 10M). The concomitant decrease in IL-17 suggests a potential shift in the cytokine milieu. Moreover, the expression of p-STAT3 was significantly elevated in the SPOCK1-high group, thereby supporting our experimental findings that SPOCK1 activates the STAT3 signaling pathway (Fig. 10N).

SPOCK1 orchestrates an immunosuppressive landscape by driving T-cell dysfunction and regulatory T-cell recruitment

To investigate the impact of SPOCK1 expression on the functional state of tumor-infiltrating T cells, we first performed bioinformatics analysis. SPOCK1 expression demonstrated a significant positive correlation with tumor immune dysfunction and exclusion (TIDE) scores (Fig. 11A). Compared to the low-expression group, the SPOCK1-high group exhibited higher TIDE and Exclusion scores, which correlate with a diminished response to immunotherapy (Fig. 11B–D). Further analysis of the TCGA-NSCLC dataset using the ssGSEA algorithm showed that the T-cell exhaustion score was significantly elevated in the SPOCK1-high group (Fig. 11E). This was accompanied by synchronized upregulation of key immune checkpoint molecules, including PDCD1 (PD-1), LAG3, TIGIT, and CTLA4 (Fig. 11F). ScRNA-seq analysis further validated this trend at the subpopulation level. The proportions of exhausted T cells (Tex) and regulatory T cells (Tregs) were significantly increased in the SPOCK1-high group (Fig. 11G, H). DEG analysis demonstrated that CD8⁺ T cells in the high-expression group were enriched with markers of exhaustion (PDCD1, HAVCR2/TIM-3, and LAG3), whereas CD4⁺ T cells exhibited distinct suppressive characteristics (high expression of FOXP3 and IL2RA/CD25) (Fig. 11I, J). At the functional level, in vitro co-culture assays confirmed that SPOCK1 significantly inhibits anti-tumor activity of the CD8⁺ T cells. Specifically, SPOCK1 overexpression impaired the cytotoxic efficacy of CD8⁺ T cells, whereas SPOCK1 knockdown enhanced the cytotoxic efficacy of CD8⁺ T cells (Fig. 11K–M). ELISA results showed that SPOCK1 overexpression significantly reduced the secretion of effector cytokines such as granzyme B (GZMB) and IFN-gamma (IFN-γ) in the co-culture supernatants, thereby demonstrating the suppressive effect of SPOCK1 on T-cell function (Fig. 11N, O). Furthermore, mIHC analysis of clinical NSCLC specimens confirmed the role of SPOCK1 in shaping the TME (Fig. 11P, Q). SPOCK1-high samples showed increased CD8⁺ T cell infiltration, higher number of PD-1⁺CD8⁺ T cells and fewer GZMB⁺CD8⁺ T cells (Fig. 11R–T). Furthermore, SPOCK1-high expression was associated with increased CD4⁺ T cell infiltration and higher proportion of FOXP3⁺CD4⁺ T cells (Tregs) (Fig. 11U, V). These results indicate that SPOCK1 overexpression promotes the infiltration of exhausted CD8⁺ T cells and Tregs into tumor tissues.

Fig. 11 — A Correlation analysis shows the relationship between TIDE scores and SPOCK1 expression in NSCLC. B TIDE analysis of the TCGA-NSCLC cohort shows significant differences in the TIDE scores between the SPOCK1-high and SPOCK1-low expression groups. C Quantification of immune exclusion scores in the SPOCK1-high and SPOCK1-low expression NSCLC cohorts. D The effect of SPOCK1 expression on the therapeutic response to immunotherapy in the TCGA-NSCLC cohort. E Quantification of T-cell exhaustion scores using ssGSEA to determine the impact of SPOCK1 expression on T-cell states using the TCGA-NSCLC dataset. F Differences in the expression levels of several T-cell exhaustion-related genes, such as PDCD1, HAVCR2, LAG3, TIGIT, and CTLA4, between the SPOCK1-high and SPOCK1-low expression groups. G UMAP visualization of T-cell sub-clusters that were re-clustered from the global scRNA-seq dataset. H Differential proportions of immune cell subsets between SPOCK1-high and SPOCK1-low samples. I Volcano plot shows differentially expressed genes (DEGs) in the CD8⁺ T cells between the SPOCK1-high and SPOCK1-low groups. Red denotes significantly upregulated genes, and blue denotes downregulated genes (criteria: P < 0.05). J Volcano plot shows the differentially expressed genes in the CD4⁺ T cells between the SPOCK1-high and SPOCK1-low NSCLC cohorts. Red denotes significantly upregulated genes, whereas significantly downregulated genes are indicated in blue. K Schematic diagram of the co-culture system involving CD8⁺ T cells and primary NSCLC cells. L, M Results for the CD8⁺ T cell-mediated NSCLC cell killing assay. Primary NSCLC cells (1 × 10⁴ cells/well) were co-cultured with autologous CD8⁺T cells for 48 h. Then, suspended T cells were removed by gentle washing. The viability of the remaining adherent tumor cells was quantified using the CCK-8 assay. The T-cell-mediated killing ratio was subsequently calculated according to the absorbance values. All experiments were performed in quadruplicate (n = 4). N, O ELISA results show the levels of GZMB and IFN-γ in the co-culture supernatants after 48 h of incubation with autologous primary NSCLC cells to evaluate the effector function of CD8⁺ T cells. P, Q Results for the mIHC assay on a tissue microarray (TMA) from NSCLC patients to characterize the immune landscape based on SPOCK1 expression levels (high vs. low). Representative images show the spatial distribution of SPOCK1 (green), CD4 (pink), CD8 (red), GZMB (yellow), PD-1 (cyan), and FOXP3 (magenta). DAPI (blue) was used as the nuclear counterstain. Quantitative analysis shows significant differences in the infiltration and functional state of immune cells between the two groups. R–V Comparative analysis of infiltrating immune cell populations in the SPOCK1-high and -low groups, including percentages of total CD8⁺ T cells (R), PD-1⁺CD8⁺ T cells (S), GZMB⁺ CD8⁺ T cells (T), total CD4⁺ T cells (U), and FOXP3⁺CD4⁺ T cells (V). The high-expression group included 34 samples, and the low-expression group included 34 samples. *indicates P < 0.05 compared to the “Low-expression group.” The scale bar represents 200 μm.

This phenotype was successfully recapitulated in a C57BL/6 mouse subcutaneous xenograft model. SPOCK1 overexpression significantly accelerated tumor growth in vivo (Supplementary Fig. 3E–G). Multiplex IHC staining of tumor tissues revealed that tumor-infiltrating T cells in the SPOCK1-overexpressing group exhibited a dysfunctional or “exhausted” phenotype, characterized by an increased proportion of PD-1⁺CD8⁺ T cells and a decreased proportion of GZMB⁺CD8⁺ T cells (Supplementary Fig. 3H–K). Furthermore, FOXP3⁺CD4⁺ T cells were also significantly increased in the SPOCK1-overexpressing group (Supplementary Fig. 3L–N).

In summary, our study demonstrates that SPOCK1 acts as a pivotal mediator in the progression of NSCLC by orchestrating an immunosuppressive microenvironment through the induction of T-cell exhaustion and promotion of Treg infiltration.

Discussion

In the study, we systematically analyzed the oncogenic role of SPOCK1 in NSCLC and its underlying pathogenetic mechanisms. Our findings demonstrate that SPOCK1 inhibits ferroptosis by modulating GSH metabolism and plays a critical role in remodeling the immunosuppressive TME. These results provide a novel perspective for understanding the malignant progression of NSCLC.

Here, we performed GWAS of a Chinese NSCLC cohort and identified SPOCK1 as a novel susceptibility gene for NSCLC. The risk locus rs4976398 showed a significant association with NSCLC, and the G allele (vs. ancestral A allele) increased disease risk by 37%. Bioinformatics analyses consistently demonstrated SPOCK1 overexpression in NSCLC compared to non-cancerous tissues. Furthermore, high SPCK1 expression correlates with advanced tumor stages and decreased patient survival. WB and IHC analyses consistently demonstrated significant upregulation of SPOCK1 protein expression in both tissue specimens and tumor cell samples from NSCLC patients. Furthermore, high SPOCK1 expression correlated with more aggressive pathological characteristics. Functional studies such as genetic silencing or ectopic overexpression of SPOCK1 demonstrated that SPOCK1 overexpression correlated with increased proliferation, migration, invasion and cell cycle progression of NSCLC cells. In vivo experiments showed that SPOCK1 ablation significantly reduced tumor growth and metastasis in xenograft models, whereas SPOCK1 overexpression promoted tumorigenesis. These observations align with previous studies implicating SPOCK1 as a driver of tumor progression in multiple cancers^9,10,66. Our results also demonstrate the role of SPOCK1 in ferroptosis resistance, a mechanism critical for tumor survival under oxidative stress.

Redox imbalance in the TME causes aberrantly high levels of ROS and contributes to tumor development^67,68. ROS plays a dual role in tumor cells. Excessive ROS levels can directly cause oxidative damage to cellular components, including DNA, proteins, and others, thereby inhibiting tumor progression; on the contrary, ROS drives metabolic reprogramming in cancer cells and promotes tumor cell survival and proliferation by activating specific signaling pathways^69–71. Increased ROS levels lead to lipid peroxidation, which in turn triggers ferroptosis in the tumor cells^72,73. GSH exerts antioxidant effects by neutralizing intracellular ROS levels and inhibiting lipid peroxidation²⁰. However, specific mechanisms regulating GSH metabolism in NSCLC progression are not clear. In this study, proteomics combined with non-targeted metabolomics revealed a novel regulatory mechanism of GSH metabolism. SPOCK1-deficient NSCLC cells exhibited significantly higher ROS and MDA levels. This exacerbated oxidative stress and significantly enhanced lipid peroxidation and ferroptosis in NSCLC cells. Concurrently, SPOCK1-deficient NSCLC cells demonstrated impaired GSH biosynthesis capacity, thereby suggesting functional impairment of the intracellular antioxidant defense system. Mechanistically, SPOCK1 inhibits ferroptosis in the tumor cells by upregulating the expression of SLC3A2 at the transcriptional level and facilitating cystine transport into cells for synthesizing GSH, the primary intracellular antioxidant. Our experiments demonstrate that SPOCK1 acts as a key driver of GSH biosynthesis in NSCLC cells, thereby regulating redox homeostasis.

STAT3 is a key dual-function oncogene that regulates both signal transduction and transcription^74,75. Transcriptional regulation, including transcriptional activation or inhibition, is a process by which specific genes are regulated through transcription factors and other molecular mechanisms during gene expression. STAT3 is a negative modulator of ferroptosis. It inhibits ferroptosis and promotes cancer progression by regulating the expression of key ferroptosis-correlated genes such as GPX4, SLC7A11, ARPC1A, and FTH1^58,59,76,77. Targeting STAT3 promotes ferroptosis in tumor cells and enhances the sensitivity of specific cancers to chemotherapy and immunotherapy^58,75. Our findings align with previous studies indicating a role for STAT3 in ferroptosis regulation. Our data suggest that SPOCK1 activates the JAK2/STAT3 signaling pathway by binding to integrin α5β1 and promoting STAT3 phosphorylation at tyrosine 705 (Tyr705) and its subsequent translocation to the nucleus. In the nucleus, STAT3 binds to the SLC3A2 promoter and enhances its transcription. Consequently, the upregulation of SLC3A2 likely supports GSH synthesis, thereby contributing to ferroptosis resistance in NSCLC cells (Fig.12).

Fig. 12 — In NSCLC, SPOCK1 is highly expressed under the transcriptional regulation of HSF1. By interacting with integrin α5β1, SPOCK1 triggers the activation of the JAK2/STAT3 signaling pathway, which subsequently upregulates SLC3A2, a key component of GSH synthesis. This regulatory axis effectively suppresses ferroptosis by bolstering antioxidant defenses, thereby promoting tumor progression. P phosphorylation.

In the treatment of NSCLC, development of resistance to chemotherapy and immunotherapy significantly limits clinical efficacy^78–80. Recent preclinical studies have demonstrated that targeting the ferroptosis regulatory pathway can effectively reverse multidrug resistance phenotypes by inducing lipid peroxidation-driven cell death in tumor cells^81,82. However, tumors of different histological origins exhibit significant heterogeneity in sensitivity to ferroptosis inducers. These differences may be linked to tissue-specific regulatory mechanisms underlying tumor cell metabolic characteristics and antioxidant defense systems. Classic ferroptosis inducers (e.g., Erastin and RSL3) disrupt tumor cell redox homeostasis by specifically inhibiting the system Xc− transporter or directly inactivating GPX4, thereby triggering iron-dependent lipid peroxide accumulation and ferroptosis^44,83,84. In this study, we demonstrated that genetic depletion of SPOCK1 increased the in vitro sensitivity of NSCLC cells to Erastin and significantly reduced tumor growth and metastasis in vivo. Furthermore, Erastin treatment of SPOCK1-deficient cells significantly increased the accumulation of 4HNE, a byproduct of lipid peroxidation, and downregulated the expression levels of SLC3A2, GPX4, and Ki67 proteins. This suggests a pivotal role of SPOCK1 in regulating ferroptosis. However, it is important to acknowledge that specific and potent small-molecule inhibitors of SPOCK1 are not yet approved for clinical use, thereby limiting immediate translational application. Consequently, our findings should be interpreted as a preclinical proof-of-concept, highlighting the SPOCK1 pathway as a promising candidate for future drug discovery. Our data suggest that targeting this pathway could potentially offer a strategy to overcome metabolic adaptations in patients with advanced or therapy-resistant NSCLC exhibiting high SPOCK1 expression. In summary, our data provide a strong rationale for developing therapies targeting the SPOCK1 signaling axis to overcome ferroptosis resistance in NSCLC cells. Therefore, future efforts should prioritize screening and optimization of small-molecule inhibitors against SPOCK1 to realize the translational potential of this combination strategy.

HSF1, a master regulator of the heat shock response, drives tumorigenesis by maintaining proteostasis and promoting stress adaptation^85–87. Building on the observed marked overexpression of SPOCK1 in NSCLC and its well-established oncogenic functions, this study further investigated the transcriptional regulatory mechanisms underlying the aberrant upregulation of SPOCK1 in NSCLC based on its high expression in NSCLC tissues and cells as well as its well-established oncogenic functions. Our experimental evidence confirmed that HSF1 directly binds to the SPOCK1 gene promoter and significantly enhances its transcriptional activity. We propose that this HSF1-driven SPOCK1 upregulation may represent a critical survival mechanism against oxidative stress. By maintaining elevated SPOCK1 levels, the HSF1-SPOCK1 axis helps protect NSCLC cells from lipid peroxidation, thereby contributing to resistance against ferroptosis.

Conventionally, a high degree of T-cell infiltration—the so-called “hot tumor” phenotype—is associated with favorable clinical prognosis^88,89. However, our study reveals a complex paradox where high SPOCK1 expression is accompanied by the recruitment of CD8⁺ and CD4⁺ T cells while simultaneously orchestrating an immunosuppressive tumor microenvironment by inducing T-cell exhaustion and Treg infiltration. This dysfunctional “high infiltration, low efficacy” state was systematically validated across multiple dimensions. Bioinformatics and scRNA-seq results demonstrated significant enrichment of Tex and Tregs in the SPOCK1-high group, as well as upregulation of inhibitory checkpoints such as PD-1, TIM-3, and FOXP3. Functionally, in vitro co-culture assays and ELISA confirmed that SPOCK1 impairs the cytotoxic potency of CD8⁺ T cells and suppresses the secretion of key effector cytokines, including IFN-γ and GZMB. Furthermore, these findings were corroborated by mouse models and mIHC analysis of NSCLC patient tissues. SPOCK1-high tumors exhibited increased infiltration of CD8⁺ T cells, primarily characterized by an expansion of the PD-1⁺ CD8⁺ T-cell subset, whereas the GZMB⁺CD8⁺ T-cell population was significantly reduced. Additionally, there was a marked increase in FOXP3⁺CD4⁺ T cells (Tregs). Taken together, these data indicate that SPOCK1 drives the immunosuppressive landscape of NSCLC by promoting T-cell exhaustion and the infiltration of Tregs.

Given that SPOCK1 also promotes ferroptosis resistance, we propose that it may reshape the metabolic landscape of TME by augmenting GSH metabolism, leading to nutrient competition or redox imbalance. This, in turn, synergistically triggers metabolic exhaustion and functional impairment of infiltrating T cells^90–92. Consequently, extensive in vitro and in vivo investigations are warranted to further elucidate the precise mechanisms by which SPOCK1 orchestrates immune evasion.

In summary, SPOCK1 represents a multifunctional oncogene that integrates genetic susceptibility, metabolic regulation, and immune evasion. By intervening in the SPOCK1-mediated inhibition of ferroptosis and its associated immunosuppressive mechanisms, novel precision therapeutic strategies may be developed for patients with advanced or treatment-resistant NSCLC.

Supplementary information

Supplementary information.^{(18.1MB, pdf)}

Acknowledgements

This work was supported by the National Natural Science Foundation of China (82472706, 82403069, 82504077), Natural Science Foundation of Jiangsu Province (BK20240488, BK20250538), Suzhou Clinical Key Disease Diagnosis and Treatment Technology Program (LCZX202339) and Research Foundation of Jingjiang People’s Hospital (JRY-KY-2025-019).

Author contributions

N.L., M.C., and K.Z. proposed and designed this study. N.L., X.Z., G.H., Y.L., M.C., and Y.L. conducted the experiments and analyzed the data. All authors participated in drafting and revising the manuscript and have submitted the final approved version to the journal.

Data availability

Data are provided within the manuscript or supplementary information files.

Competing interests

The authors declare no competing interests.

Footnotes

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

These authors contributed equally: Na Liu, Xiao-Ren Zhu, Yang Liu.

Contributor Information

Na Liu, Email: 2697075944@qq.com.

Min-Bin Chen, Email: cmb1981@163.com.

Supplementary information

The online version contains supplementary material available at https://doi.org/10.1038/s41698-026-01455-5.

References

1.Hendriks, L. E. L. et al. Non-small-cell lung cancer. Nat. Res. Dis. Prim.10, 71 (2024). [DOI] [PubMed] [Google Scholar]
2.Ren, F. et al. Liquid biopsy techniques and lung cancer: diagnosis, monitoring and evaluation. J. Exp. Clin. Cancer Res.43, 96 (2024). [DOI] [PMC free article] [PubMed]
3.Cooper, A. J., Sequist, L. V. & Lin, J. J. Third-generation EGFR and ALK inhibitors: mechanisms of resistance and management. Nat. Rev. Clin. Oncol.19, 499–514 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Pike, L. R. G. et al. Tyrosine kinase inhibitors with and without up-front stereotactic radiosurgery for brain metastases from EGFR and ALK oncogene-driven non-small cell lung cancer (TURBO-NSCLC). J. Clin. Oncol.42, 3606–3617 (2024). [DOI] [PMC free article] [PubMed]
5.Zhao, Y. et al. Efficacy and safety of immune checkpoint inhibitors for individuals with advanced EGFR-mutated non-small-cell lung cancer who progressed on EGFR tyrosine-kinase inhibitors: a systematic review, meta-analysis, and network meta-analysis. Lancet Oncol.25, 1347–1356 (2024). [DOI] [PubMed] [Google Scholar]
6.Qin, B. D. et al. Immunotherapy-based regimens for patients with EGFR-mutated non-small cell lung cancer who progressed on EGFR-TKI therapy. J. Immunother. Cancer12 (2024). [DOI] [PMC free article] [PubMed]
7.Váncza, L., Tátrai, P., Reszegi, A., Baghy, K. & Kovalszky, I. SPOCK1 with unexpected function. The start of a new career. Am. J. Physiol. Cell Physiol.322, C688–c693 (2022). [DOI] [PubMed] [Google Scholar]
8.Xiao, M., Xue, J. & Jin, E. SPOCK: master regulator of malignant tumors (Review). Mol. Med. Rep.30 (2024). [DOI] [PMC free article] [PubMed]
9.Shu, Y. J. et al. SPOCK1 as a potential cancer prognostic marker promotes the proliferation and metastasis of gallbladder cancer cells by activating the PI3K/AKT pathway. Mol. Cancer14, 12 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Chien, M. H. et al. Targeting the SPOCK1-snail/slug axis-mediated epithelial-to-mesenchymal transition by apigenin contributes to repression of prostate cancer metastasis. J. Exp. Clin. Cancer Res.38, 246 (2019). [DOI] [PMC free article] [PubMed]
11.Yu, F. et al. SPOCK1 is upregulated in recurrent glioblastoma and contributes to metastasis and Temozolomide resistance. Cell Prolif.49, 195–206 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Babaei-Jadidi, R. et al. Anti-miR-135/SPOCK1 axis antagonizes the influence of metabolism on drug response in intestinal/colon tumour organoids. Oncogenesis11, 4 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Li, H. et al. RBM15 facilitates osimertinib resistance of lung adenocarcinoma through m6A-dependent epigenetic silencing of SPOCK1. Oncogene44, 307–321 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Jiang, X., Stockwell, B. R. & Conrad, M. Ferroptosis: mechanisms, biology and role in disease. Nat. Rev. Mol. Cell Biol.22, 266–282 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Dixon, S. J. et al. Ferroptosis: an iron-dependent form of nonapoptotic cell death. Cell149, 1060–1072 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Yan, Y. et al. SLC7A11 expression level dictates differential responses to oxidative stress in cancer cells. Nat. Commun.14, 3673 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Zhang, W. et al. RBMS1 regulates lung cancer ferroptosis through translational control of SLC7A11. J. Clin. Invest.131 (2021). [DOI] [PMC free article] [PubMed]
18.Yang, W. S. et al. Regulation of ferroptotic cancer cell death by GPX4. Cell156, 317–331 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Klein Geltink, R. I., O’Sullivan, D. & Pearce, E. L. Caught in the cROSsfire: GSH controls T cell metabolic reprogramming. Immunity46, 525–527 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Niu, B. et al. Application of glutathione depletion in cancer therapy: enhanced ROS-based therapy, ferroptosis, and chemotherapy. Biomaterials277, 121110 (2021). [DOI] [PubMed] [Google Scholar]
21.Ogiwara, H. et al. Targeting the vulnerability of glutathione metabolism in ARID1A-deficient cancers. Cancer Cell35, 177–190.e178 (2019). [DOI] [PubMed] [Google Scholar]
22.Cheng, X., Xu, H. D., Ran, H. H., Liang, G. & Wu, F. G. Glutathione-depleting nanomedicines for synergistic cancer therapy. ACS Nano15, 8039–8068 (2021). [DOI] [PubMed] [Google Scholar]
23.Wang, Y. et al. Neuronal-driven glioma growth requires Gαi1 and Gαi3. Theranostics11, 8535–8549 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Gao, Y. Y. et al. The histone acetyltransferase HBO1 functions as a novel oncogenic gene in osteosarcoma. Theranostics11, 4599–4615 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Yao, J. et al. The requirement of phosphoenolpyruvate carboxykinase 1 for angiogenesis in vitro and in vivo. Sci. Adv.8, eabn6928 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Zhu, X. R. et al. A novel identified epithelial ligand-receptor-associated gene signature highlights POPDC3 as a potential therapy target for non-small cell lung cancer. Cell Death Dis.16, 114 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Liu, Y. Y. et al. microRNA-200a downregulation in human glioma leads to Gαi1 over-expression, Akt activation, and cell proliferation. Oncogene37, 2890–2902 (2018). [DOI] [PubMed] [Google Scholar]
28.Zhu, X. R. et al. Identification of phosphoenolpyruvate carboxykinase 1 as a potential therapeutic target for pancreatic cancer. Cell Death Dis.12, 918 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Jiang, J. Z. et al. Identification of Gαi3 as a promising molecular oncotarget of pancreatic cancer. Cell Death Dis.15, 699 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Peng, S. Q. et al. Identification of matrix-remodeling associated 5 as a possible molecular oncotarget of pancreatic cancer. Cell Death Dis.14, 157 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Shi, J. et al. Genome-wide association study of lung adenocarcinoma in East Asia and comparison with a European population. Nat. Commun.14, 3043 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Tian, Z. et al. CAS Array: design and assessment of a genotyping array for Chinese biobanking. Precis. Clin. Med.6, pbad002 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Xie, T. et al. Cohort profile: kunshan aging research with E-health (KARE). Int. J. Epidemiol.54 (2025). [DOI] [PubMed]
34.Liu, Y. et al. Association of mitochondrial DNA copy number with chronic kidney disease in older adults. BMC Geriatr.23, 514 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Gong, J. et al. PancanQTL: systematic identification of cis-eQTLs and trans-eQTLs in 33 cancer types. Nucleic Acids Res.46, D971–d976 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Dowling, C. M., Zhang, H., Chonghaile, T. N. & Wong, K. K. Shining a light on metabolic vulnerabilities in non-small cell lung cancer. Biochim. Biophys. Acta Rev. Cancer1875, 188462 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Lv, L., Huang, R. H., Li, J., Xu, J. & Gao, W. Impact of NSCLC metabolic remodeling on immunotherapy effectiveness. Biomark. Res.10, 66 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Bajinka, O., Ouedraogo, S. Y., Golubnitschaja, O., Li, N. & Zhan, X. Energy metabolism as the hub of advanced non-small cell lung cancer management: a comprehensive view in the framework of predictive, preventive, and personalized medicine. EPMA J.15, 289–319 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Morrissey, S. M. et al. Tumor-derived exosomes drive immunosuppressive macrophages in a pre-metastatic niche through glycolytic dominant metabolic reprogramming. Cell Metab.33, 2040–2058.e2010 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
40.Hao, B. et al. Identification of SLC2A1 as a predictive biomarker for survival and response to immunotherapy in lung squamous cell carcinoma. Comput. Biol. Med.171, 108183 (2024). [DOI] [PubMed] [Google Scholar]
41.Slieker, R. C. et al. Identification of biomarkers for glycaemic deterioration in type 2 diabetes. Nat. Commun.14, 2533 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
42.Alshargabi, R. et al. SPOCK1 induces adipose tissue maturation: New insights into the function of SPOCK1 in metabolism. Biochem. Biophys. Res. Commun.533, 1076–1082 (2020). [DOI] [PubMed] [Google Scholar]
43.Liu, Y. et al. The diversified role of mitochondria in ferroptosis in cancer. Cell Death Dis.14, 519 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Zhou, Q. et al. Ferroptosis in cancer: from molecular mechanisms to therapeutic strategies. Signal Transduct. Target. Ther.9, 55 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
45.Jin, Y. et al. Targeting polarized phenotype of microglia via IL6/JAK2/STAT3 signaling to reduce NSCLC brain metastasis. Signal Transduct. Target. Ther.7, 52 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
46.Liang, L. et al. Autophagy inhibition potentiates the anti-angiogenic property of multikinase inhibitor anlotinib through JAK2/STAT3/VEGFA signaling in non-small cell lung cancer cells. J. Exp. Clin. Cancer Res.38, 71 (2019) [DOI] [PMC free article] [PubMed]
47.Wang, L. N. et al. TGF-β1/SH2B3 axis regulates anoikis resistance and EMT of lung cancer cells by modulating JAK2/STAT3 and SHP2/Grb2 signaling pathways. Cell Death Dis.13, 472 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
48.O’Brown, N. M. et al. The secreted neuronal signal Spock1 promotes blood-brain barrier development. Dev. Cell58, 1534–1547.e1536 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
49.Ye, Z. et al. SPOCK1: a multi-domain proteoglycan at the crossroads of extracellular matrix remodeling and cancer development. Am. J. Cancer Res.10, 3127–3137 (2020). [PMC free article] [PubMed] [Google Scholar]
50.Kanchanawong, P. & Calderwood, D. A. Organization, dynamics and mechanoregulation of integrin-mediated cell-ECM adhesions. Nat. Rev. Mol. Cell Biol.24, 142–161 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
51.Chastney, M. R., Kaivola, J., Leppänen, V. M. & Ivaska, J. The role and regulation of integrins in cell migration and invasion. Nat. Rev. Mol.Cell Biol.26, 147–167 (2025). [DOI] [PubMed] [Google Scholar]
52.Du, Z. et al. SPOCK1 overexpression induced by platelet-derived growth factor-BB promotes hepatic stellate cell activation and liver fibrosis through the integrin α5β1/PI3K/Akt signaling pathway. Lab. Invest.100, 1042–1056 (2020). [DOI] [PubMed] [Google Scholar]
53.Mitra, A. K. et al. Ligand-independent activation of c-Met by fibronectin and α(5)β(1)-integrin regulates ovarian cancer invasion and metastasis. Oncogene30, 1566–1576 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
54.Chen, Z. et al. ANGPTL3 promotes colorectal carcinoma progression and metastasis through regulating COL1A2 transcription via interacting with integrin αVβ3. Cancer Lett.638, 218150 (2026). [DOI] [PubMed] [Google Scholar]
55.Wu, Q. et al. Inhibition of bladder cancer growth with homoharringtonine by inactivating integrin α5/β1-FAK/Src axis: a novel strategy for drug application. Pharmacol. Res.188, 106654 (2023). [DOI] [PubMed] [Google Scholar]
56.Yang, X. et al. FAP promotes immunosuppression by cancer-associated fibroblasts in the tumor microenvironment via STAT3-CCL2 signaling. Cancer Res.76, 4124–4135 (2016). [DOI] [PubMed] [Google Scholar]
57.Pan, Y. M. et al. STAT3 signaling drives EZH2 transcriptional activation and mediates poor prognosis in gastric cancer. Mol. Cancer15, 79 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
58.Ouyang, S. et al. Inhibition of STAT3-ferroptosis negative regulatory axis suppresses tumor growth and alleviates chemoresistance in gastric cancer. Redox Biol.52, 102317 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
59.Lai, J. Q. et al. Baicalein triggers ferroptosis in colorectal cancer cells via blocking the JAK2/STAT3/GPX4 axis. Acta Pharmacol. Sin.45, 1715–1726 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
60.Wang, L. et al. ATF3 promotes erastin-induced ferroptosis by suppressing system Xc(). Cell Death Differ.27, 662–675 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
61.Lin, A., Wei, T., Meng, H., Luo, P. & Zhang, J. Role of the dynamic tumor microenvironment in controversies regarding immune checkpoint inhibitors for the treatment of non-small cell lung cancer (NSCLC) with EGFR mutations. Mol. Cancer18, 139 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
62.Mollaoglu, G. et al. The lineage-defining transcription factors SOX2 and NKX2-1 determine lung cancer cell fate and shape the tumor immune microenvironment. Immunity49, 764–779.e769 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
63.Cascone, T. et al. Tumor immunology and immunotherapy of non-small-cell lung cancer. Cold Spring Harb. Perspect. Med.12 (2022) [DOI] [PMC free article] [PubMed]
64.Lin, H. et al. Itaconate transporter SLC13A3 impairs tumor immunity via endowing ferroptosis resistance. Cancer Cell42, 2032–2044.e2036 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
65.Ping, Y. et al. PD-1 signaling limits expression of phospholipid phosphatase 1 and promotes intratumoral CD8(+) T cell ferroptosis. Immunity57, 2122–2139.e2129 (2024). [DOI] [PubMed] [Google Scholar]
66.Perurena, N. et al. EPCR promotes breast cancer progression by altering SPOCK1/testican 1-mediated 3D growth. J. Hematol. Oncol.10, 23 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
67.Sabharwal, S. S. & Schumacker, P. T. Mitochondrial ROS in cancer: initiators, amplifiers or an Achilles’ heel? Nat. Rev. Cancer14, 709–721 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
68.Cui, Q. et al. Modulating ROS overcome multidrug resistance cancer. Drug Resist. Updat.41, 1–25 (2018). [DOI] [PubMed] [Google Scholar]
69.Wang, K. et al. Targeting metabolic-redox circuits for cancer therapy. Trends Biochem. Sci.44, 401–414 (2019). [DOI] [PubMed] [Google Scholar]
70.Wang, L. et al. Melatonin and erastin emerge synergistic anti-tumor effects on oral squamous cell carcinoma by inducing apoptosis, ferroptosis, and inhibiting autophagy through promoting ROS. Cell. Mol. Biol. Lett.28, 36 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
71.Yang, H. et al. The role of cellular reactive oxygen species in cancer chemotherapy. J. Exp. Clin. Cancer Res.37, 266 (2018). [DOI] [PMC free article] [PubMed]
72.Park, E. & Chung, S. W. ROS-mediated autophagy increases intracellular iron levels and ferroptosis by ferritin and transferrin receptor regulation. Cell Death Dis.10, 822 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
73.Zhang, S. et al. Chemotherapy impairs ovarian function through excessive ROS-induced ferroptosis. Cell Death Dis.14, 340 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
74.Yu, H., Lee, H., Herrmann, A., Buettner, R. & Jove, R. Revisiting STAT3 signalling in cancer: new and unexpected biological functions. Nat. Rev. Cancer14, 736–746 (2014). [DOI] [PubMed] [Google Scholar]
75.Zou, S. et al. Targeting STAT3 in cancer immunotherapy. Mol. cancer19, 145 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
76.Wu, Y. et al. Ginsenoside Rh3 induces pyroptosis and ferroptosis through the Stat3/p53/NRF2 axis in colorectal cancer cells. Acta Biochim. Biophys. Sin.55, 587–600 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
77.Ji, J., Li, H., Wang, W., Yuan, B. & Shen, T. ARPC1A is regulated by STAT3 to inhibit ferroptosis and promote prostate cancer progression. Hum. Cell35, 1591–1601 (2022). [DOI] [PubMed] [Google Scholar]
78.Ernst, S. M. et al. Utilizing ctDNA to discover mechanisms of resistance to targeted therapies in patients with metastatic NSCLC: towards more informative trials. Nat. Rev. Clin. Oncol.22, 371–378 (2025). [DOI] [PubMed] [Google Scholar]
79.Meyer, M. L. et al. New promises and challenges in the treatment of advanced non-small-cell lung cancer. Lancet404, 803–822 (2024). [DOI] [PubMed] [Google Scholar]
80.Mariniello, A. et al. Primary and acquired resistance to immunotherapy with checkpoint inhibitors in nsclc: from bedside to bench and back. BioDrugs39, 215–235 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
81.Zhang, C., Liu, X., Jin, S., Chen, Y. & Guo, R. Ferroptosis in cancer therapy: a novel approach to reversing drug resistance. Mol. Cancer21, 47 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
82.Wang, Y. et al. Overcoming cancer chemotherapy resistance by the induction of ferroptosis. Drug Resist. Updat.66, 100916 (2023) [DOI] [PubMed]
83.Ghoochani, A. et al. Ferroptosis inducers are a novel therapeutic approach for advanced prostate cancer. Cancer Res.81, 1583–1594 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
84.Sun, S., Shen, J., Jiang, J., Wang, F. & Min, J. Targeting ferroptosis opens new avenues for the development of novel therapeutics. Signal Transduct. Target. Ther.8, 372 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
85.Huang, M. et al. HSF1 facilitates the multistep process of lymphatic metastasis in bladder cancer via a novel PRMT5-WDR5-dependent transcriptional program. Cancer Commun.42, 447–470 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
86.Chin, Y. et al. Targeting HSF1 for cancer treatment: mechanisms and inhibitor development. Theranostics13, 2281–2300 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
87.Pariollaud, M. et al. Circadian disruption enhances HSF1 signaling and tumorigenesis in Kras-driven lung cancer. Sci. Adv.8, eabo1123 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
88.Liu, Y. T. et al. Turning cold tumors into hot tumors to ignite immunotherapy. Mol. Cancer24, 254 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
89.Galon, J. & Bruni, D. Approaches to treat immune hot, altered and cold tumours with combination immunotherapies. Nat. Rev. Drug Discov.18, 197–218 (2019). [DOI] [PubMed] [Google Scholar]
90.Yu, T. et al. Inhibition of glutamate-to-glutathione flux promotes tumor antigen presentation in colorectal cancer cells. Adv. Sci.12, e2310308 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
91.Kurniawan, H. et al. Glutathione restricts serine metabolism to preserve regulatory T cell function. Cell Metab.31, 920–936.e927 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
92.Rivadeneira, D. B. et al. Oxidative-stress-induced telomere instability drives T cell dysfunction in cancer. Immunity58, 2524–2540.e2525 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary information.^{(18.1MB, pdf)}

Data Availability Statement

Data are provided within the manuscript or supplementary information files.

[CR1] 1.Hendriks, L. E. L. et al. Non-small-cell lung cancer. Nat. Res. Dis. Prim.10, 71 (2024). [DOI] [PubMed] [Google Scholar]

[CR2] 2.Ren, F. et al. Liquid biopsy techniques and lung cancer: diagnosis, monitoring and evaluation. J. Exp. Clin. Cancer Res.43, 96 (2024). [DOI] [PMC free article] [PubMed]

[CR3] 3.Cooper, A. J., Sequist, L. V. & Lin, J. J. Third-generation EGFR and ALK inhibitors: mechanisms of resistance and management. Nat. Rev. Clin. Oncol.19, 499–514 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR4] 4.Pike, L. R. G. et al. Tyrosine kinase inhibitors with and without up-front stereotactic radiosurgery for brain metastases from EGFR and ALK oncogene-driven non-small cell lung cancer (TURBO-NSCLC). J. Clin. Oncol.42, 3606–3617 (2024). [DOI] [PMC free article] [PubMed]

[CR5] 5.Zhao, Y. et al. Efficacy and safety of immune checkpoint inhibitors for individuals with advanced EGFR-mutated non-small-cell lung cancer who progressed on EGFR tyrosine-kinase inhibitors: a systematic review, meta-analysis, and network meta-analysis. Lancet Oncol.25, 1347–1356 (2024). [DOI] [PubMed] [Google Scholar]

[CR6] 6.Qin, B. D. et al. Immunotherapy-based regimens for patients with EGFR-mutated non-small cell lung cancer who progressed on EGFR-TKI therapy. J. Immunother. Cancer12 (2024). [DOI] [PMC free article] [PubMed]

[CR7] 7.Váncza, L., Tátrai, P., Reszegi, A., Baghy, K. & Kovalszky, I. SPOCK1 with unexpected function. The start of a new career. Am. J. Physiol. Cell Physiol.322, C688–c693 (2022). [DOI] [PubMed] [Google Scholar]

[CR8] 8.Xiao, M., Xue, J. & Jin, E. SPOCK: master regulator of malignant tumors (Review). Mol. Med. Rep.30 (2024). [DOI] [PMC free article] [PubMed]

[CR9] 9.Shu, Y. J. et al. SPOCK1 as a potential cancer prognostic marker promotes the proliferation and metastasis of gallbladder cancer cells by activating the PI3K/AKT pathway. Mol. Cancer14, 12 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR10] 10.Chien, M. H. et al. Targeting the SPOCK1-snail/slug axis-mediated epithelial-to-mesenchymal transition by apigenin contributes to repression of prostate cancer metastasis. J. Exp. Clin. Cancer Res.38, 246 (2019). [DOI] [PMC free article] [PubMed]

[CR11] 11.Yu, F. et al. SPOCK1 is upregulated in recurrent glioblastoma and contributes to metastasis and Temozolomide resistance. Cell Prolif.49, 195–206 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR12] 12.Babaei-Jadidi, R. et al. Anti-miR-135/SPOCK1 axis antagonizes the influence of metabolism on drug response in intestinal/colon tumour organoids. Oncogenesis11, 4 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR13] 13.Li, H. et al. RBM15 facilitates osimertinib resistance of lung adenocarcinoma through m6A-dependent epigenetic silencing of SPOCK1. Oncogene44, 307–321 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR14] 14.Jiang, X., Stockwell, B. R. & Conrad, M. Ferroptosis: mechanisms, biology and role in disease. Nat. Rev. Mol. Cell Biol.22, 266–282 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR15] 15.Dixon, S. J. et al. Ferroptosis: an iron-dependent form of nonapoptotic cell death. Cell149, 1060–1072 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR16] 16.Yan, Y. et al. SLC7A11 expression level dictates differential responses to oxidative stress in cancer cells. Nat. Commun.14, 3673 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR17] 17.Zhang, W. et al. RBMS1 regulates lung cancer ferroptosis through translational control of SLC7A11. J. Clin. Invest.131 (2021). [DOI] [PMC free article] [PubMed]

[CR18] 18.Yang, W. S. et al. Regulation of ferroptotic cancer cell death by GPX4. Cell156, 317–331 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR19] 19.Klein Geltink, R. I., O’Sullivan, D. & Pearce, E. L. Caught in the cROSsfire: GSH controls T cell metabolic reprogramming. Immunity46, 525–527 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR20] 20.Niu, B. et al. Application of glutathione depletion in cancer therapy: enhanced ROS-based therapy, ferroptosis, and chemotherapy. Biomaterials277, 121110 (2021). [DOI] [PubMed] [Google Scholar]

[CR21] 21.Ogiwara, H. et al. Targeting the vulnerability of glutathione metabolism in ARID1A-deficient cancers. Cancer Cell35, 177–190.e178 (2019). [DOI] [PubMed] [Google Scholar]

[CR22] 22.Cheng, X., Xu, H. D., Ran, H. H., Liang, G. & Wu, F. G. Glutathione-depleting nanomedicines for synergistic cancer therapy. ACS Nano15, 8039–8068 (2021). [DOI] [PubMed] [Google Scholar]

[CR23] 23.Wang, Y. et al. Neuronal-driven glioma growth requires Gαi1 and Gαi3. Theranostics11, 8535–8549 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR24] 24.Gao, Y. Y. et al. The histone acetyltransferase HBO1 functions as a novel oncogenic gene in osteosarcoma. Theranostics11, 4599–4615 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR25] 25.Yao, J. et al. The requirement of phosphoenolpyruvate carboxykinase 1 for angiogenesis in vitro and in vivo. Sci. Adv.8, eabn6928 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR26] 26.Zhu, X. R. et al. A novel identified epithelial ligand-receptor-associated gene signature highlights POPDC3 as a potential therapy target for non-small cell lung cancer. Cell Death Dis.16, 114 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR27] 27.Liu, Y. Y. et al. microRNA-200a downregulation in human glioma leads to Gαi1 over-expression, Akt activation, and cell proliferation. Oncogene37, 2890–2902 (2018). [DOI] [PubMed] [Google Scholar]

[CR28] 28.Zhu, X. R. et al. Identification of phosphoenolpyruvate carboxykinase 1 as a potential therapeutic target for pancreatic cancer. Cell Death Dis.12, 918 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR29] 29.Jiang, J. Z. et al. Identification of Gαi3 as a promising molecular oncotarget of pancreatic cancer. Cell Death Dis.15, 699 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR30] 30.Peng, S. Q. et al. Identification of matrix-remodeling associated 5 as a possible molecular oncotarget of pancreatic cancer. Cell Death Dis.14, 157 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR31] 31.Shi, J. et al. Genome-wide association study of lung adenocarcinoma in East Asia and comparison with a European population. Nat. Commun.14, 3043 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR32] 32.Tian, Z. et al. CAS Array: design and assessment of a genotyping array for Chinese biobanking. Precis. Clin. Med.6, pbad002 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR33] 33.Xie, T. et al. Cohort profile: kunshan aging research with E-health (KARE). Int. J. Epidemiol.54 (2025). [DOI] [PubMed]

[CR34] 34.Liu, Y. et al. Association of mitochondrial DNA copy number with chronic kidney disease in older adults. BMC Geriatr.23, 514 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR35] 35.Gong, J. et al. PancanQTL: systematic identification of cis-eQTLs and trans-eQTLs in 33 cancer types. Nucleic Acids Res.46, D971–d976 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR36] 36.Dowling, C. M., Zhang, H., Chonghaile, T. N. & Wong, K. K. Shining a light on metabolic vulnerabilities in non-small cell lung cancer. Biochim. Biophys. Acta Rev. Cancer1875, 188462 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR37] 37.Lv, L., Huang, R. H., Li, J., Xu, J. & Gao, W. Impact of NSCLC metabolic remodeling on immunotherapy effectiveness. Biomark. Res.10, 66 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR38] 38.Bajinka, O., Ouedraogo, S. Y., Golubnitschaja, O., Li, N. & Zhan, X. Energy metabolism as the hub of advanced non-small cell lung cancer management: a comprehensive view in the framework of predictive, preventive, and personalized medicine. EPMA J.15, 289–319 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR39] 39.Morrissey, S. M. et al. Tumor-derived exosomes drive immunosuppressive macrophages in a pre-metastatic niche through glycolytic dominant metabolic reprogramming. Cell Metab.33, 2040–2058.e2010 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR40] 40.Hao, B. et al. Identification of SLC2A1 as a predictive biomarker for survival and response to immunotherapy in lung squamous cell carcinoma. Comput. Biol. Med.171, 108183 (2024). [DOI] [PubMed] [Google Scholar]

[CR41] 41.Slieker, R. C. et al. Identification of biomarkers for glycaemic deterioration in type 2 diabetes. Nat. Commun.14, 2533 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR42] 42.Alshargabi, R. et al. SPOCK1 induces adipose tissue maturation: New insights into the function of SPOCK1 in metabolism. Biochem. Biophys. Res. Commun.533, 1076–1082 (2020). [DOI] [PubMed] [Google Scholar]

[CR43] 43.Liu, Y. et al. The diversified role of mitochondria in ferroptosis in cancer. Cell Death Dis.14, 519 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR44] 44.Zhou, Q. et al. Ferroptosis in cancer: from molecular mechanisms to therapeutic strategies. Signal Transduct. Target. Ther.9, 55 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR45] 45.Jin, Y. et al. Targeting polarized phenotype of microglia via IL6/JAK2/STAT3 signaling to reduce NSCLC brain metastasis. Signal Transduct. Target. Ther.7, 52 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR46] 46.Liang, L. et al. Autophagy inhibition potentiates the anti-angiogenic property of multikinase inhibitor anlotinib through JAK2/STAT3/VEGFA signaling in non-small cell lung cancer cells. J. Exp. Clin. Cancer Res.38, 71 (2019) [DOI] [PMC free article] [PubMed]

[CR47] 47.Wang, L. N. et al. TGF-β1/SH2B3 axis regulates anoikis resistance and EMT of lung cancer cells by modulating JAK2/STAT3 and SHP2/Grb2 signaling pathways. Cell Death Dis.13, 472 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR48] 48.O’Brown, N. M. et al. The secreted neuronal signal Spock1 promotes blood-brain barrier development. Dev. Cell58, 1534–1547.e1536 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR49] 49.Ye, Z. et al. SPOCK1: a multi-domain proteoglycan at the crossroads of extracellular matrix remodeling and cancer development. Am. J. Cancer Res.10, 3127–3137 (2020). [PMC free article] [PubMed] [Google Scholar]

[CR50] 50.Kanchanawong, P. & Calderwood, D. A. Organization, dynamics and mechanoregulation of integrin-mediated cell-ECM adhesions. Nat. Rev. Mol. Cell Biol.24, 142–161 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR51] 51.Chastney, M. R., Kaivola, J., Leppänen, V. M. & Ivaska, J. The role and regulation of integrins in cell migration and invasion. Nat. Rev. Mol.Cell Biol.26, 147–167 (2025). [DOI] [PubMed] [Google Scholar]

[CR52] 52.Du, Z. et al. SPOCK1 overexpression induced by platelet-derived growth factor-BB promotes hepatic stellate cell activation and liver fibrosis through the integrin α5β1/PI3K/Akt signaling pathway. Lab. Invest.100, 1042–1056 (2020). [DOI] [PubMed] [Google Scholar]

[CR53] 53.Mitra, A. K. et al. Ligand-independent activation of c-Met by fibronectin and α(5)β(1)-integrin regulates ovarian cancer invasion and metastasis. Oncogene30, 1566–1576 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR54] 54.Chen, Z. et al. ANGPTL3 promotes colorectal carcinoma progression and metastasis through regulating COL1A2 transcription via interacting with integrin αVβ3. Cancer Lett.638, 218150 (2026). [DOI] [PubMed] [Google Scholar]

[CR55] 55.Wu, Q. et al. Inhibition of bladder cancer growth with homoharringtonine by inactivating integrin α5/β1-FAK/Src axis: a novel strategy for drug application. Pharmacol. Res.188, 106654 (2023). [DOI] [PubMed] [Google Scholar]

[CR56] 56.Yang, X. et al. FAP promotes immunosuppression by cancer-associated fibroblasts in the tumor microenvironment via STAT3-CCL2 signaling. Cancer Res.76, 4124–4135 (2016). [DOI] [PubMed] [Google Scholar]

[CR57] 57.Pan, Y. M. et al. STAT3 signaling drives EZH2 transcriptional activation and mediates poor prognosis in gastric cancer. Mol. Cancer15, 79 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR58] 58.Ouyang, S. et al. Inhibition of STAT3-ferroptosis negative regulatory axis suppresses tumor growth and alleviates chemoresistance in gastric cancer. Redox Biol.52, 102317 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR59] 59.Lai, J. Q. et al. Baicalein triggers ferroptosis in colorectal cancer cells via blocking the JAK2/STAT3/GPX4 axis. Acta Pharmacol. Sin.45, 1715–1726 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR60] 60.Wang, L. et al. ATF3 promotes erastin-induced ferroptosis by suppressing system Xc(). Cell Death Differ.27, 662–675 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR61] 61.Lin, A., Wei, T., Meng, H., Luo, P. & Zhang, J. Role of the dynamic tumor microenvironment in controversies regarding immune checkpoint inhibitors for the treatment of non-small cell lung cancer (NSCLC) with EGFR mutations. Mol. Cancer18, 139 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR62] 62.Mollaoglu, G. et al. The lineage-defining transcription factors SOX2 and NKX2-1 determine lung cancer cell fate and shape the tumor immune microenvironment. Immunity49, 764–779.e769 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR63] 63.Cascone, T. et al. Tumor immunology and immunotherapy of non-small-cell lung cancer. Cold Spring Harb. Perspect. Med.12 (2022) [DOI] [PMC free article] [PubMed]

[CR64] 64.Lin, H. et al. Itaconate transporter SLC13A3 impairs tumor immunity via endowing ferroptosis resistance. Cancer Cell42, 2032–2044.e2036 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR65] 65.Ping, Y. et al. PD-1 signaling limits expression of phospholipid phosphatase 1 and promotes intratumoral CD8(+) T cell ferroptosis. Immunity57, 2122–2139.e2129 (2024). [DOI] [PubMed] [Google Scholar]

[CR66] 66.Perurena, N. et al. EPCR promotes breast cancer progression by altering SPOCK1/testican 1-mediated 3D growth. J. Hematol. Oncol.10, 23 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR67] 67.Sabharwal, S. S. & Schumacker, P. T. Mitochondrial ROS in cancer: initiators, amplifiers or an Achilles’ heel? Nat. Rev. Cancer14, 709–721 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR68] 68.Cui, Q. et al. Modulating ROS overcome multidrug resistance cancer. Drug Resist. Updat.41, 1–25 (2018). [DOI] [PubMed] [Google Scholar]

[CR69] 69.Wang, K. et al. Targeting metabolic-redox circuits for cancer therapy. Trends Biochem. Sci.44, 401–414 (2019). [DOI] [PubMed] [Google Scholar]

[CR70] 70.Wang, L. et al. Melatonin and erastin emerge synergistic anti-tumor effects on oral squamous cell carcinoma by inducing apoptosis, ferroptosis, and inhibiting autophagy through promoting ROS. Cell. Mol. Biol. Lett.28, 36 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR71] 71.Yang, H. et al. The role of cellular reactive oxygen species in cancer chemotherapy. J. Exp. Clin. Cancer Res.37, 266 (2018). [DOI] [PMC free article] [PubMed]

[CR72] 72.Park, E. & Chung, S. W. ROS-mediated autophagy increases intracellular iron levels and ferroptosis by ferritin and transferrin receptor regulation. Cell Death Dis.10, 822 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR73] 73.Zhang, S. et al. Chemotherapy impairs ovarian function through excessive ROS-induced ferroptosis. Cell Death Dis.14, 340 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR74] 74.Yu, H., Lee, H., Herrmann, A., Buettner, R. & Jove, R. Revisiting STAT3 signalling in cancer: new and unexpected biological functions. Nat. Rev. Cancer14, 736–746 (2014). [DOI] [PubMed] [Google Scholar]

[CR75] 75.Zou, S. et al. Targeting STAT3 in cancer immunotherapy. Mol. cancer19, 145 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR76] 76.Wu, Y. et al. Ginsenoside Rh3 induces pyroptosis and ferroptosis through the Stat3/p53/NRF2 axis in colorectal cancer cells. Acta Biochim. Biophys. Sin.55, 587–600 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR77] 77.Ji, J., Li, H., Wang, W., Yuan, B. & Shen, T. ARPC1A is regulated by STAT3 to inhibit ferroptosis and promote prostate cancer progression. Hum. Cell35, 1591–1601 (2022). [DOI] [PubMed] [Google Scholar]

[CR78] 78.Ernst, S. M. et al. Utilizing ctDNA to discover mechanisms of resistance to targeted therapies in patients with metastatic NSCLC: towards more informative trials. Nat. Rev. Clin. Oncol.22, 371–378 (2025). [DOI] [PubMed] [Google Scholar]

[CR79] 79.Meyer, M. L. et al. New promises and challenges in the treatment of advanced non-small-cell lung cancer. Lancet404, 803–822 (2024). [DOI] [PubMed] [Google Scholar]

[CR80] 80.Mariniello, A. et al. Primary and acquired resistance to immunotherapy with checkpoint inhibitors in nsclc: from bedside to bench and back. BioDrugs39, 215–235 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR81] 81.Zhang, C., Liu, X., Jin, S., Chen, Y. & Guo, R. Ferroptosis in cancer therapy: a novel approach to reversing drug resistance. Mol. Cancer21, 47 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR82] 82.Wang, Y. et al. Overcoming cancer chemotherapy resistance by the induction of ferroptosis. Drug Resist. Updat.66, 100916 (2023) [DOI] [PubMed]

[CR83] 83.Ghoochani, A. et al. Ferroptosis inducers are a novel therapeutic approach for advanced prostate cancer. Cancer Res.81, 1583–1594 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR84] 84.Sun, S., Shen, J., Jiang, J., Wang, F. & Min, J. Targeting ferroptosis opens new avenues for the development of novel therapeutics. Signal Transduct. Target. Ther.8, 372 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR85] 85.Huang, M. et al. HSF1 facilitates the multistep process of lymphatic metastasis in bladder cancer via a novel PRMT5-WDR5-dependent transcriptional program. Cancer Commun.42, 447–470 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR86] 86.Chin, Y. et al. Targeting HSF1 for cancer treatment: mechanisms and inhibitor development. Theranostics13, 2281–2300 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR87] 87.Pariollaud, M. et al. Circadian disruption enhances HSF1 signaling and tumorigenesis in Kras-driven lung cancer. Sci. Adv.8, eabo1123 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR88] 88.Liu, Y. T. et al. Turning cold tumors into hot tumors to ignite immunotherapy. Mol. Cancer24, 254 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR89] 89.Galon, J. & Bruni, D. Approaches to treat immune hot, altered and cold tumours with combination immunotherapies. Nat. Rev. Drug Discov.18, 197–218 (2019). [DOI] [PubMed] [Google Scholar]

[CR90] 90.Yu, T. et al. Inhibition of glutamate-to-glutathione flux promotes tumor antigen presentation in colorectal cancer cells. Adv. Sci.12, e2310308 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR91] 91.Kurniawan, H. et al. Glutathione restricts serine metabolism to preserve regulatory T cell function. Cell Metab.31, 920–936.e927 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR92] 92.Rivadeneira, D. B. et al. Oxidative-stress-induced telomere instability drives T cell dysfunction in cancer. Immunity58, 2524–2540.e2525 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

SPOCK1 acts as a negative regulator of ferroptosis in non-small cell lung cancer

Na Liu

Xiao-Ren Zhu

Yang Liu

Guo-Hu Han

Yuan-Yuan Liu

Kai-Xin Zhou

Min-Bin Chen

Abstract

Introduction

Methods

Reagents and antibodies

Genome-wide association study

Cell culture, clinical specimens, and tissue microarray

Single-cell RNA sequencing data and cell–cell communication analysis

Untargeted metabolomics

Plasmids, siRNAs, lentiviral construction, and cell transfection

Cellular functional studies and gene/protein detection

Immunofluorescence and immunohistochemical staining

Multiplex immunohistochemistry

In vitro ferroptosis marker assays

Transmission electron microscopy (TEM) assays

Co-immunoprecipitation (Co-IP)

Chromatin immunoprecipitation

Dual-luciferase reporter assays

Electrophoretic mobility shift assay

Evaluation of CD8 + T-cell-induced tumor cytotoxicity

Animal experiments

Subcutaneous xenograft studies in mice

Lung metastasis assay

Statistical analysis

Results

Expression and prognostic relevance of SPOCK1 in NSCLC

Fig. 1. Expression and prognostic relevance of SPOCK1 in non-small cell lung cancer.

SPOCK1 silencing induces robust anti-NSCLC cell activity

Fig. 2. SPOCK1 silencing induces robust anti-NSCLC cell activity.

Overexpression of SPOCK1 promotes tumorigenic activity in NSCLC cells

Fig. 3. SPOCK1 overexpression promotes tumorigenic activity in NSCLC cells.

Proteomic and metabolomic analyses link SPOCK1 to ferroptosis regulation in NSCLC

Fig. 4. Proteomic and metabolomic analyses link SPOCK1 to ferroptosis regulation in NSCLC.

SPOCK1 silencing promotes ferroptosis in NSCLC cells

Fig. 5. SPOCK1 silencing promotes ferroptosis in NSCLC cells.

SPOCK1 regulates GSH biosynthesis by upregulating SLC3A2

Fig. 6. SPOCK1 regulates GSH biosynthesis by upregulating SLC3A2.

SPOCK1 upregulates SLC3A2 expression by activating the JAK2/STAT3 pathway

Fig. 7. SPOCK1 upregulates SLC3A2 expression by activating the JAK2/STAT3 pathway.

SPOCK1 silencing enhances erastin-induced ferroptosis synergistically

Fig. 8. SPOCK1 silencing enhances erastin-induced ferroptosis.

HSF1 is a master transcription factor of SPOCK1 in NSCLC cells

Fig. 9. HSF1 is a master transcription factor of SPOCK1 in NSCLC cells.

The expression of SPOCK1 in the NSCLC microenvironment is associated with the extent of immune cell infiltration

Fig. 10. The expression of SPOCK1 in the NSCLC microenvironment is associated with the extent of immune cell infiltration.

SPOCK1 orchestrates an immunosuppressive landscape by driving T-cell dysfunction and regulatory T-cell recruitment

Fig. 11. SPOCK1 orchestrates an immunosuppressive landscape by driving T-cell dysfunction and regulatory T-cell recruitment.

Discussion

Fig. 12. Schematic diagram shows the mechanism by which SPOCK1 regulates NSCLC progression.

Supplementary information

Acknowledgements

Author contributions

Data availability

Competing interests

Footnotes

Contributor Information

Supplementary information

References

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases