Bioinformatics analysis and experimental validation confirm COL6A3 as a promising target for renal cell carcinoma therapy

Introduction

Worldwide, kidney cancer is a prevalent urological malignancy, ranking as the sixth most common cancer in men and the tenth in women (1,2). According to GLOBOCAN 2020 statistics, there were approximately 431,288 new cases of kidney cancer and 179,368 deaths globally, with a higher incidence and mortality observed in men compared to women (3-5). Despite advances in medical care, the incidence rate continues to rise rapidly. Histologically, clear cell renal cell carcinoma (ccRCC), which originates in the proximal tubule, is the most common subtype, accounting for approximately 70–90% of cases, followed by papillary renal cell carcinoma (RCC) (10–15%) (6). Although surgical interventions—such as nephron-sparing surgery, radical nephrectomy, and partial nephrectomy—remain the primary treatments for RCC, the patient mortality rate remains high (7-10). In recent years, immunotherapy has revolutionized the therapeutic paradigm for advanced RCC. Specifically, immune checkpoint inhibitors (ICIs) targeting the programmed death 1 (PD-1)/programmed death-ligand 1 (PD-L1) and cytotoxic T-lymphocyte-associated antigen 4 (CTLA-4) pathways have demonstrated remarkable clinical effectiveness, significantly improving objective response rates and overall survival (OS) (11,12). Despite these breakthroughs, the broader application of current immunotherapies is hindered by prominent limitations. A substantial proportion of patients experience primary resistance and fail to respond to initial treatments, while many initial responders eventually develop acquired resistance (13). Furthermore, the lack of reliable predictive biomarkers for patient stratification and the potential for severe immune-related adverse events remain major clinical challenges (14). Therefore, there is an urgent need to deeply explore the RCC immune microenvironment and identify novel therapeutic targets to overcome these clinical hurdles. Therefore, there is an urgent need to identify novel therapeutic targets for RCC.

Collagen VI alpha 3 (COL6A3) is one of the three alpha chains of type VI collagen (alongside COL6A1 and COL6A2). Previous studies have demonstrated that COL6A3 expression is positively correlated with body mass index (BMI) and adipose tissue content (15-18). Interestingly, beyond its role in metabolic conditions like diabetes and obesity, COL6A3 plays a pivotal role in tumorigenesis and cancer development of osteosarcoma, colorectal cancer, meningiomas, and triple-negative breast cancer (19-23). It is noteworthy that COL6A3 appears to be associated with the PI3K-AKT pathway, a key pathway regulating tumor progression. For example, previous studies have demonstrated that silencing COL6A3 can not only inhibit tumor progression but also suppress the PI3K-AKT pathway in gastric cancer and osteosarcoma (20,24). In addition, a non-tumor study found that when the Avellanin A was treated with prostate hyperplasia cells, the expression of COL6A3 and PI3K-AKT pathway-related genes were reduced (25). However, the specific role of COL6A3 and the relationship between COL6A3 and PI3K-AKT pathways in kidney cancer remains largely unexplored. This study aims to investigate the involvement of COL6A3 in the occurrence, development, prognosis, and immune landscape of renal cancer through bioinformatics analysis and in vitro and in vivo experiments. We present this article in accordance with the MDAR and ARRIVE reporting checklists (available at https://tau.amegroups.com/article/view/10.21037/tau-2026-0295/rc).

Methods

R packages and database

The relationship between COL6A3 and the prognosis of RCC was analyzed through the Gene Expression Profiling Interactive Analysis 2 (GEPIA2) database (http://gepia2.cancer-pku.cn/#analysis), ‘survival’ package and ‘survminer’ package. Patients were divided into high- and low-expression groups according to the median expression level of the target gene. OS was estimated using the Kaplan-Meier method, and differences between groups were compared using the log-rank test. Through University of Alabama at Birmingham Cancer Data Analysis Portal (UALCAN) database [UALCAN (https://www.uab.edu/)], we analyzed the relationship between COL6A3 expression and RCC grade, lymph node metastasis and TP53 mutation. Gene expression differences among clinicopathological subgroups, were analyzed using UALCAN, in which statistical significance was estimated by Welch’s t-test. RCC transcriptome data were obtained from The Cancer Genome Atlas (TCGA) database (https://cancergenome.nih.gov/) and utilized for validation of the difference analysis and paired difference analysis with ‘ggpubr’ and ‘limma’ packages. The expression level of the target gene was log₂-transformed as log₂(expression +1). Differences between tumor and normal tissues were compared using the Wilcoxon rank-sum test. Paired tumor and adjacent normal samples from the same patients were compared using the paired Wilcoxon signed-rank test. The Tumor Immune Estimation Resource (TIMER) database [TIMER2.0 (https://compbio.cn/timer2/)] was used to verify the expression difference of COL6A3 in various normal tissue samples and cancer tissue samples. Differences between tumor and normal tissues were compared using the Wilcoxon rank-sum test. Subsequently, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis was performed using the clusterProfiler R package. Differentially expressed genes (DEGs) were converted to Entrez IDs and subjected to over-representation analysis. Enriched GO terms were identified using the hypergeometric test, and multiple testing correction was performed using the Benjamini-Hochberg method. Terms with P value <0.05 and q value <0.05 were considered statistically significant. In addition, we scored immune and mesenchymal COL6A3 in RCC via the ‘limma’ package, the ‘estimate’ package and the ‘reshape2’ package. Immune cell infiltration fractions were estimated using CIBERSORT, and samples with CIBERSORT P value <0.05 were retained. Differences in immune cell fractions between the high- and low-expression groups were compared using the Wilcoxon rank-sum test. Subsequently, the relationship between COL6A3 and immune cell infiltration was revealed by ‘limma’. Finally, the relationship between COL6A3 and immune checkpoints was verified by exploration through ‘ggplot2’ and ‘reshape2’ packages (P value filter set to 0.001). Tumor samples were divided into high- and low-expression groups according to the median expression level of the target gene. Differences in responsiveness to ICIs between patients with high and low expression of COL6A3 were analyzed by immune proportion score (IPS)-CTLA-4 blockers and IPS-PD-1/PD-L1 blockers data from The Cancer Imaging Archive (TCIA) (https://tcia.at). Differences in TCIA-derived immunotherapy-related scores between the two groups were compared using the Wilcoxon rank-sum test. We obtained external independent immune validation cohort data from the Gene Expression Omnibus (GEO) database (https://www.ncbi.nlm.nih.gov/geo/; GSE67501) and applied the ‘pRRophetic’ package to predict drug sensitivity of COL6A3. Tumor samples were divided into high- and low-expression groups according to the median expression level of the target gene. Differences in predicted half-maximal inhibitory concentration (IC₅₀) values between the two groups were compared using the Wilcoxon rank-sum test. Drugs with P value <0.001 were considered significantly different.

Cell culture

Human RCC cell lines, 786-O (Procell, CL-0010, Wuhan, China) and 769-P (Procell, CL-0009), were maintained in a humidified incubator at 37 ℃ with 5% CO₂. The standard growth medium utilized was RPMI 1640 (BasalMedia, Shanghai, China), which was enriched with 10% fetal bovine serum (FBS; HAKATA, Shanghai, China) and 1% penicillin-streptomycin mixture (BasalMedia).

Chemicals and antibodies

Cell Counting Kit-8 (CCK-8) was acquired in APExBIO (Shanghai, China). Radioimmunoprecipitation assay (RIPA), phenylmethanesulfonyl fluoride (PMSF), paraformaldehyde, and crystal violet desalination solution were obtained from Solarbio (Beijing, China). The enhanced chemiluminescence (ECL) detection reagent was sourced from ProteinTech (Wuhan, China), while TRIzol reagent was acquired from Ambion (Thermo Fisher Scientific, Austin, TX, USA). Matrixgel was purchased from Corning Inc. (Corning, NY, USA). Anti-COL6A3 (K112951P) was purchased from Solarbio. Anti-PI3K (20584-1-AP), anti-AKT (60203-2-Ig), anti-phosphorylated-AKT (p-AKT) (Ser473) (28731-1-AP), anti-glyceraldehyde 3-phosphate dehydrogenase (anti-GAPDH) (10494-1-AP), and other secondary antibodies were obtained from ProteinTech.

Western blot assay

Following the removal of culture media, cells underwent two PBS washes before being lysed in a pre-mixed buffer of RIPA and PMSF (100:1 ratio) for 60 minutes at 4 ℃. The resulting lysates were centrifuged at 12,000 rpm for 20 minutes to harvest the soluble protein fractions. After electrophoretic separation, proteins were blotted onto polyvinylidene difluoride (PVDF) membranes (Merck Millipore, Carrigtwohill, Ireland). To prevent non-specific binding, the membranes were blocked with 5% skim milk, followed by an overnight incubation at 4 ℃ with the designated primary antibodies. Subsequently, the blots were probed with appropriate secondary antibodies. Finally, immunoreactive bands were visualized via ECL reagents and captured by a Tanon chemiluminescence imaging system (Shanghai, China).

Quantitative real-time polymerase chain reaction (qRT-PCR) and messenger RNA (mRNA) extraction

Following RNA isolation using TRIzol reagent, complementary DNA (cDNA) was synthesized utilizing the BIOG cDNA Synthesis Kit (BioDai, Changzhou, China). qRT-PCR was subsequently conducted on an ABI Prism 7900 sequence detection system (Thermo Fisher Scientific, Waltham, MA, USA). To determine relative changes in gene expression, the 2^−ΔΔCt method was applied, with GAPDH serving as the endogenous normalization control. The following primers were used: GAPDH-forward: 5'-AATGGGCAGCCGTTAGGAAA-3'; GAPDH-reverse: 5'-GCGCCCAATACGACCAAATC-3'; COL6A3-forward: 5'-AAATGGTGCGGCTGCTGATA-3'; COL6A3-reverse: 5'-CAAGGCCATCCTTCGAGTGT-3'.

CCK-8 assay

To assess cell proliferation over a 5-day period, cells seeded in 96-well plates were treated with CCK-8 reagent at designated time points and incubated for 2 hours. Subsequently, the optical density (OD) at 450 nm was quantified utilizing a full-wavelength scanning microplate reader (BioTek, Montpelier, VT, USA). Data acquisition and visualization were performed employing Gen5 software (BioTek; v1.11.5).

Colony formation assay

To assess clonogenic ability, cells were seeded into six-well plates at a density of 1×10³ cells per well. Following approximately 14 days of incubation, the culture medium was discarded. The resulting colonies were then fixed with paraformaldehyde for 20 minutes and subsequently stained using crystal violet for an additional 20 minutes. Finally, images of the colonies were captured using a digital camera (Canon, Tokyo, Japan).

5-ethynyl-2'-deoxyuridine (EdU) assay

To evaluate cell proliferation, 5×10⁴ cells from various groups were seeded into 24-well plates and incubated in complete medium at 37 ℃ for 24 hours. The cells were then labeled utilizing an EdU assay kit (Abbkine, Shanghai, China). Following incubation, samples were fixed with 4% paraformaldehyde and subsequently permeabilized using 1% Triton X-100 (Solarbio). Nuclei were counterstained with Hoechst (Abbkine) for 10 minutes, and the resulting fluorescence signals were captured employing a fluorescence microscope (Olympus, Tokyo, Japan).

Wound scratch assay

Different cells were seeded in six-well plates. Cells were scratched with a 200 µL pipette tip and photographed under microscope at 0, 24, and 48 hours (Olympus).

Transwell assays

First, spread Matrixgel on the bottom of the transwell chamber. Then, 600 µL of full medium was added to the lower layer of the small chamber, and 200 µL of FBS-free medium was added to the upper part of the small chamber. A total of 5×10⁴ cells were added to the upper layer of the chamber. After 48 hours of incubation, cells were fixed with paraformaldehyde for 15 minutes and stained with crystal violet for 15 minutes. The cells in the upper layer of the chamber were wiped by cotton swab. Finally, photos were taken under a microscope (Olympus).

Tumor xenografts in nude mice

A total of 10 6-week-old male BALB/c nude mice, procured from Yishengyuan (Tianjin, China), were maintained in an Assessment and Accreditation of Laboratory Animal Care International (AAALAC)-accredited animal facility at Tianjin Medical University under a controlled environment (temperature-regulated, 12-hour light/dark cycle). The animals were randomly assigned to either the negative control (NC) or sh-COL6A3 cohort. For xenograft establishment, 1×10⁶ cells from the corresponding groups were injected subcutaneously into the groin region of each mouse. Following a 14-day growth period after initial tumor formation, the animals were euthanized via cervical dislocation. The excised tumors were subsequently weighed, and their volumes were calculated using the formula: V = 0.5 × length × width². All animal experiments were performed under a project license (No. 2023006) granted by the Ethics Committee of Tianjin Medical University, in compliance with institutional guidelines for the care and use of animals.

Immunohistochemistry (IHC) staining

Tissue specimens were sourced from the Department of Pathology at The Second Hospital of Tianjin Medical University. IHC was conducted following the manufacturer’s protocol using a commercial IHC kit (ProteinTech). Briefly, sections were sequentially deparaffinized, rehydrated, and subjected to antigen retrieval, followed by the quenching of endogenous peroxidase activity. To prevent non-specific binding, slides were blocked with goat serum and subsequently incubated overnight at 4 ℃ with primary antibodies against COL6A3 (1:400), Ki67 (1:400), and vimentin (1:4,000). Following intermediate PBS washes, the sections were treated with a reaction enhancer and an enzyme-conjugated goat anti-rabbit/mouse immunoglobulin G (IgG) polymer. Immunoreactivity was visualized using 3,3'-diaminobenzidine (DAB), after which the slides were counterstained with hematoxylin, dehydrated, and mounted. The staining intensity of tumor cells was evaluated on a three-point scale (1, weak; 2, moderate; 3, strong). Finally, image acquisition was performed utilizing an optical microscope (Olympus). Experiments involving human tissue were approved by the Ethics Committee of The Second Hospital of Tianjin Medical University (approval No. KS2024029). This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. Informed consent was obtained from all patients whose tissues were used.

Hematoxylin and eosin (H&E) staining analysis

H&E staining was conducted utilizing a commercial kit (Solarbio; Cat. No. G1120). Briefly, tissue sections were deparaffinized with xylene and rehydrated through an ethanol gradient. The slides were then immersed in hematoxylin for 3 minutes. Following a deionized water wash, the eosin solution was applied for 1 minute. Finally, the samples were dehydrated, mounted with neutral resin, and imaged under an optical microscope (Olympus).

Statistical analysis

Statistical analyses were conducted utilizing GraphPad Prism software (GraphPad Software, San Diego, CA, USA) (v10.6). All quantitative results are expressed as the mean ± standard deviation (SD). To evaluate differences between two independent groups, an unpaired, two-tailed Student’s t-test was applied. For experimental designs involving comparisons among three or more groups, a one-way analysis of variance (ANOVA) was implemented. Bioinformatics results were analyzed for differences using the log-rank test, Welch’s t-test, Wilcoxon rank-sum test, and hypergeometric test. P<0.05 was considered to be statistically significant (*, P<0.05; **, P<0.01; ***, P<0.001) and all tests were two-tailed.

Results

Correlation of COL6A3 expression with RCC clinicopathology and prognosis

To assess the prognostic value of COL6A3 in RCC, we interrogated its correlation with patient outcomes using the GEPIA2 platform. Kaplan-Meier analysis revealed that high COL6A3 expression portended dismal disease-free survival (DFS), progression-free survival (PFS), and OS (Figure 1A-1C). Concurrently, transcript levels of COL6A3 climbed with advancing tumor grade and clinical stage (Figure 1D,1E) and closely mirrored the extent of lymph node metastasis (Figure 1F). Notably, this gene was found to be differentially overexpressed in various RCC subtypes relative to normal counterparts (Figure 1G). Collectively, these data implicate COL6A3 as a key driver of tumor progression and metastasis, positioning it as a viable prognostic indicator and a targetable liability for RCC intervention.

Figure 1 The association between COL6A3 and prognosis and clinical progression. (A) The relationship between COL6A3 and DFS. (B) The relationship between COL6A3 and PFS. (C) The relationship between COL6A3 and OS. (D) Expression of COL6A3 in KIRC based on tumor grade. (E) Expression of COL6A3 in KIRC based on individual cancer grade. (F) Expression of COL6A3 in KIRC based on nodal metastasis status. (G) Expression of COL6A3 in KIRC based on KIRC subtypes. All data were expressed as the mean ± SD. The log-rank test and Welch’s t-test were used to assess statistical differences. *, P<0.05; **, P<0.01; ***, P<0.001; ns, not significant. COL6A3, collagen VI alpha 3; DFS, disease-free survival; HR, hazard ratio; KIRC, kidney renal clear cell carcinoma; OS, overall survival; PFS, progression-free survival; SD, standard deviation; TCGA, The Cancer Genome Atlas.

COL6A3 is highly expressed in RCC at tissue and cellular level

To robustly validate COL6A3 enrichment in RCC, we employed a dual empirical and computational approach. Protein and mRNA assessments of matched clinical specimens revealed a stark accumulation of COL6A3 in malignant tissues via Western blot and qRT-PCR (Figure 2A,2B). This overexpression pattern was further visualized in situ through IHC staining of 20 patient pairs (Figure 2C). In vitro modeling faithfully recapitulated these clinical observations, demonstrating profound transcript and protein amplification across a panel of RCC cell lines (786-O, A-498, ACHN, and 769-P) relative to normal HK-2 cells (Figure 2D,2E). Orthogonal validation using TCGA cohorts independently confirmed this differential upregulation (Figure 2F,2G). Extending this paradigm, TIMER-based pan-cancer mapping underscored a conserved elevation of this molecule across multiple neoplastic contexts, prominently including renal clear cell carcinoma (Figure 2H). Together, these multi-tiered datasets unequivocally establish COL6A3 as a highly expressed molecular signature in RCC.

Figure 2 Expression of COL6A3 in tissues and cell lines. (A) Expression of COL6A3 protein in adjacent tissues and cancer tissues. GAPDH was used as an internal reference. (B) Expression of COL6A3 mRNA in adjacent tissues and RCC tissues. (C) IHC staining of adjacent tissues and RCC tissues. Scale bar =100 µm. (D) Expression of COL6A3 protein in different RCC cell lines. (E) Expression of COL6A3 mRNA in different RCC cell lines. (F) The difference in COL6A3 expression between normal renal tissue and renal cancer by using the TCGA database. (G) The paired difference analysis in COL6A3 expression between normal renal tissue and renal cancer by using the TCGA database. (H) The expression differences of COL6A3 in normal tissues and tumor tissues in each system. Quantitative data are presented as the mean ± SD. To determine statistical significance between two independent groups, an unpaired, two-tailed Student’s t-test was employed. For experimental designs involving three or more groups, a one-way ANOVA was utilized. Bioinformatics results were compared using the Wilcoxon rank-sum test. *, P<0.05; **, P<0.01; ***, P<0.001; ****, P<0.0001. ANOVA, analysis of variance; COL6A3, collagen VI alpha 3; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; IHC, immunohistochemistry; mRNA, messenger RNA; RCC, renal cell carcinoma; SD, standard deviation; TCGA, The Cancer Genome Atlas.

Functional enrichment analysis of COL6A3

To elucidate the molecular networks orchestrated by COL6A3, we stratified tumor samples based on its abundance and conducted differential transcriptomic profiling. The distinct gene expression signatures defining the high-COL6A3 and low-COL6A3 cohorts were captured via hierarchical clustering (Figure 3A). Subsequent functional annotation utilizing GO and KEGG frameworks mapped these DEGs predominantly to the PI3K-AKT signaling cascade and immune regulatory processes (Figure 3B,3C). Ultimately, this computational evidence establishes a strong mechanistic rationale for our downstream empirical investigations into the immunomodulatory properties of COL6A3.

Figure 3 Function analysis of COL6A3. (A) Grouped variance analysis of COL6A3. (B) KEGG enrichment analysis. (C) GO enrichment analysis. All data were expressed as mean ± SD. The Wilcoxon rank-sum test and hypergeometric test were used to assess statistical differences. Multiple testing correction was performed using the Benjamini-Hochberg method. BP, biological process; CC, cellular component; COL6A3, collagen VI alpha 3; GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes; MF, molecular function; SD, standard deviation.

Silencing of COL6A3 effectively inhibits RCC in vitro through PI3K-AKT pathway

To determine if COL6A3 drives RCC malignancy via the PI3K-AKT axis, we executed loss-of-function studies. Immunoblotting confirmed that COL6A3 depletion in 786-O and 769-P cells effectively abrogated the expression of PI3K and p-AKT, without altering total AKT pools (Figure 4A). Phenotypically, COL6A3 silencing profoundly blunted tumor cell expansion. This was evidenced by curtailed viability in CCK-8 analyses, diminished clonogenic survival and stalled DNA synthesis during EdU incorporation (Figure 4B-4D). Furthermore, the metastatic potential of these cells was severely compromised; both wound closure dynamics and Transwell invasion rates plummeted upon COL6A3 knockdown (Figure 4E,4F; quantified in Figure S1). Together, these data phenocopy our bioinformatic predictions, establishing that COL6A3 facilitates RCC proliferation and metastasis, likely by engaging the PI3K-AKT signaling cascade.

Figure 4 The effects of COL6A3 on the biological behavior of RCC. (A) Knockdown of COL6A3 and changes in PI3K-AKT pathway-related proteins. (B) Effect of knockdown of COL6A3 on RCC cell lines activity by CCK-8 assay. (C) Effect of knockdown of COL6A3 on proliferation of RCC cell lines in clone formation assays (crystal violet staining). (D) The effect of decreased COL6A3 on mitosis in RCC cell lines. (E) Effect of knockdown of COL6A3 on migratory capacity of RCC cell lines in wound healing assay. (F) Effect of knockdown of COL6A3 on the invasive capacity of RCC cell lines in Transwell assays (crystal violet staining). Quantitative data are presented as the mean ± SD, n=3. To determine statistical significance between two independent groups, an unpaired, two-tailed Student’s t-test was employed. CCK-8, Cell Counting Kit-8; COL6A3, collagen VI alpha 3; EdU, 5-ethynyl-2'-deoxyuridine; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; NC, negative control; p-, phosphorylated-; RCC, renal cell carcinoma; SD, standard deviation; sh, short hairpin RNA.

Silencing COL6A3 inhibits the growth of RCC in vivo

Building upon our in vitro findings, we established a murine subcutaneous xenograft model to evaluate the in vivo oncogenic role of COL6A3. Following 14 days of engraftment, macroscopic examination of the resected tissues revealed that COL6A3 depletion severely restricted tumor expansion, yielding significantly lower tumor volumes and weights compared to the NC group (Figure 5A-5C). Histological profiling further substantiated these physical observations; IHC demonstrated a marked downregulation of both Ki67 (a proliferation indicator) and vimentin (an invasiveness marker) in the COL6A3-silenced lesions (Figure 5D). Importantly, H&E staining of major visceral organs confirmed that this targeted genetic knockdown induced no observable systemic toxicity (Figure 5E). Concordant with our cellular assays, these in vivo data unequivocally validate COL6A3 as a critical driver of RCC progression and underscore its viability as a highly promising therapeutic target.

Figure 5 In vivo evaluation of COL6A3 on tumor growth. (A) Representative images of the excised xenograft tumors from the indicated groups. (B) Tumor volume growth curves (calculated as V = 0.5 × length × width²) for the two cohorts. (C) Quantitative analysis of final tumor weights. (D) Representative H&E and IHC staining images evaluating tumor-associated biomarkers in the xenograft tissues. (E) Histological assessment (H&E staining) of major vital organs, including the heart, liver, spleen, lungs, and kidneys of the nude mice. Quantitative data are presented as the mean ± SD, n=5. To determine statistical significance between two independent groups, an unpaired, two-tailed Student’s t-test was employed. *, P<0.05; **, P<0.01. COL6A3, collagen VI alpha 3; H&E, hematoxylin and eosin; IHC, immunohistochemistry; NC, negative control; SD, standard deviation; sh, short hairpin RNA.

Differential analysis of immune cell infiltration of COL6A3

To delineate the potential immunomodulatory role of COL6A3 in RCC, we profiled the tumor immune microenvironment. Infiltration analysis across 22 distinct leukocyte subsets revealed that anti-tumor effector cells—specifically CD8⁺ T cells and activated natural killer (NK) cells—were significantly enriched in the low-COL6A3 cohort (Figure 6A). Building upon these distinct immune landscapes, we utilized computational models to predict the therapeutic response to ICIs. To provide a comprehensive assessment, we categorized the predicted ICI responses into four distinct treatment subtypes based on the applied blockades: (I) neither CTLA-4 nor PD-1 blockers (CTLA-4⁻/PD-1⁻); (II) PD-1 blockade alone (CTLA-4⁻/PD-1⁺); (III) CTLA-4 blockade alone (CTLA-4⁺/PD-1⁻); and (IV) combination therapy utilizing both blockades (CTLA-4⁺/PD-1⁺). These in silico analyses suggested that patients with low COL6A3 expression exhibited superior responsiveness across the active treatment modalities, particularly benefiting from anti-PD-1, anti-CTLA-4, and their combination (Figure 6B). Collectively, these predictive findings imply that high COL6A3 expression correlates with an immunosuppressive state. While empirical and clinical validations are necessary, our data raise the hypothesis that targeting COL6A3 could serve as a strategy to modulate the tumor microenvironment, potentially enhancing the efficacy of immunotherapies in RCC.

Figure 6 Analysis of the effect of COL6A3 on the immune microenvironment and immunotherapy value of COL6A3. (A) Immune cells infiltration analysis of COL6A3. (B) Sensitivity analysis of four subtypes of ICIs anti-PD-1 and anti-CTLA-4. All data were expressed as mean ± SD. The Wilcoxon rank-sum test was used to assess statistical differences. *, P<0.05; **, P<0.01; ***, P<0.001. COL6A3, collagen VI alpha 3; CTLA-4, cytotoxic T-lymphocyte-associated antigen 4; ICI, immune checkpoint inhibitor; NK, natural killer; PD-1, programmed death 1; SD, standard deviation.

Prediction of COL6A3 responsiveness to targeted drugs and chemotherapeutic agents

To evaluate the potential of COL6A3 in guiding clinical interventions, we assessed chemotherapeutic sensitivities using the ‘pRRophetic’ algorithm. While high COL6A3 expression correlated with diminished sensitivity to erlotinib (Figure 7A), this same cohort exhibited heightened susceptibility to a broad spectrum of targeted agents, including dasatinib, bexarotene, imatinib, rapamycin, sunitinib, paclitaxel, and pazopanib (Figure 7B-7H). Ultimately, these data highlight the utility of COL6A3 as a predictive biomarker for patient stratification, paving the way for more tailored and effective therapies in RCC.

Figure 7 Chemotherapeutic drug sensitivity analysis of COL6A3. (A) Erlotinib. (B) Dasatinib. (C) Bexarotene. (D) Imatinib. (E) Rapamycin. (F) Sunitinib. (G) Paclitaxel. (H) Pazopanib. All data were expressed as mean ± SD. The Wilcoxon rank-sum test was used to assess statistical differences. *, P<0.05; **, P<0.01; ***, P<0.001. COL6A3, collagen VI alpha 3; IC₅₀, half-maximal inhibitory concentration; SD, standard deviation.

Discussion

RCC remains the most prevalent malignancy of the urinary system, characterized by high mortality rates despite advances in surgical and chemotherapeutic interventions (2,4). Consequently, the identification of novel biomarkers for early diagnosis and targeted therapy is urgently needed. While COL6A3 has been implicated in metabolic disorders and tumorigenesis of gastric cancer and osteosarcoma via the PI3K-AKT pathway, its specific role in RCC has remained largely unexplored (19,20,24). In the current study, we integrated bioinformatics analysis with experimental validation to elucidate the diagnostic, prognostic, and therapeutic value of COL6A3 in RCC.

Our investigation revealed that COL6A3 is significantly upregulated in RCC tissues and cell lines, a finding corroborated by TCGA and TIMER database analyses. Clinically, elevated COL6A3 expression was positively correlated with advanced tumor grade, stage, lymph node metastasis, and poor prognosis, suggesting its potential as a robust prognostic biomarker. Mechanistically, we demonstrated that COL6A3 exerts its oncogenic effects by activating the PI3K-AKT signaling pathway. Silencing COL6A3 significantly downregulated key pathway components and impaired RCC cell viability, proliferation, and metastatic potential (migration and invasion). This finding is consistent with previous studies on the effects of COL6A3 on the PI3K-AKT pathway (20,24,26,27). However, this study is the first to demonstrate in the RCC that silencing COL6A3 inhibits the PI3K-AKT pathway, a key pathway for tumor progression.

Currently, immunotherapy—particularly ICIs—has fundamentally transformed the therapeutic landscape for advanced RCC, demonstrating remarkable clinical effectiveness. However, its widespread success is still restricted by significant limitations, including heterogeneous patient response rates, primary or acquired resistance, and a critical scarcity of reliable predictive biomarkers. Addressing these challenges requires a deeper understanding of the tumor immune microenvironment. Beyond its oncogenic role, our study highlights the potential implications of COL6A3 within the tumor immune microenvironment. Interestingly, low COL6A3 expression was associated with higher immune and stromal scores, suggesting a more immunologically active state. Specifically, we observed a computational enrichment of CD8⁺ T cells and activated NK cells in the low-expression group (28-32). Furthermore, our predictive models estimate that RCC patients with high COL6A3 levels might be less responsive to ICIs treatments compared to those with low expression. Consequently, while targeting COL6A3 to alleviate immunosuppression presents an intriguing theoretical strategy, these predictions necessitate rigorous experimental and clinical validation. Nevertheless, these in silico findings offer a conceptual framework for evaluating COL6A3 as a predictive biomarker for immunotherapy in RCC.

In addition to immunotherapeutic implications, our study evaluated the impact of COL6A3 on conventional chemotherapeutic and targeted drug sensitivities. Using the ‘pRRophetic’ algorithm, we observed that high COL6A3 expression was associated with diminished sensitivity to erlotinib yet exhibited a striking theoretical vulnerability to a panel of alternative agents, including standard-of-care tyrosine kinase inhibitors (TKIs) for RCC such as sunitinib and pazopanib, as well as the mTOR inhibitor Rapamycin. The predicted sensitivity to rapamycin is particularly noteworthy. Given our experimental in vitro and in vivo findings that COL6A3 activates the PI3K-AKT signaling cascade, it is biologically plausible that high-COL6A3 tumors develop an oncogenic dependency on this pathway, thereby rendering them highly susceptible to downstream mTOR inhibition. These findings suggest that profiling COL6A3 status could facilitate precision medicine by identifying patients most likely to benefit from specific TKIs or PI3K/AKT/mTOR axis inhibitors.

Despite these promising findings, our study has certain limitations. First, while we validated COL6A3 expression using clinical samples from our center, the sample size was relatively small, and larger cohorts are needed to confirm its diagnostic sensitivity. Second, although we provided extensive in vitro and in vivo evidence of COL6A3’s oncogenic role, our conclusions regarding its impact on immunotherapy response rely primarily on bioinformatic predictions. Future studies utilizing immunocompetent animal models or clinical trial cohorts are necessary to empirically validate these immunomodulatory effects.

Conclusions

In conclusion, our study identifies COL6A3 as a novel prognostic biomarker and a promising therapeutic vulnerability in RCC, particularly through its involvement in the PI3K-AKT pathway and its potential regulatory role in tumor immunity.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the MDAR and ARRIVE reporting checklists. Available at https://tau.amegroups.com/article/view/10.21037/tau-2026-0295/rc

Data Sharing Statement: Available at https://tau.amegroups.com/article/view/10.21037/tau-2026-0295/dss

Peer Review File: Available at https://tau.amegroups.com/article/view/10.21037/tau-2026-0295/prf

Funding: This work was supported by the Spark Scientific Research Project of The First Hospital of Hebei Medical University (No. XH202514), the Medical Science Research Project of Hebei (No. 20250372), and the Hebe Province Government Funded Clinical Medicine Excellent Talents Training Program (No. ZF2026069).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tau.amegroups.com/article/view/10.21037/tau-2026-0295/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. All animal experiments were performed under a project license (No. 2023006) granted by the Ethics Committee of Tianjin Medical University, in compliance with institutional guidelines for the care and use of animals. Experiments involving human tissue were approved by the Ethics Committee of The Second Hospital of Tianjin Medical University (Approval No. KS2024029). This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. Informed consent was obtained from all patients whose tissues were used.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Bahadoram S, Davoodi M, Hassanzadeh S, et al. Renal cell carcinoma: an overview of the epidemiology, diagnosis, and treatment. G Ital Nefrol 2022;39:2022-vol3.
2. Miller KD, Goding Sauer A, Ortiz AP, et al. Cancer Statistics for Hispanics/Latinos, 2018. CA Cancer J Clin 2018;68:425-45. [Crossref] [PubMed]
3. Sung H, Ferlay J, Siegel RL, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin 2021;71:209-49. [Crossref] [PubMed]
4. Turčić M, Krpina K, Trivanović D, et al. Rethinking Advanced Renal Cell Carcinoma: Integrative Genomics, Immunotherapy, and Molecular-Orthomolecular Strategies. Cancers (Basel) 2026;18:1435. [Crossref] [PubMed]
5. Bray F, Ferlay J, Soerjomataram I, et al. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2018;68:394-424. [Crossref] [PubMed]
6. Warren AY, Harrison D. WHO/ISUP classification, grading and pathological staging of renal cell carcinoma: standards and controversies. World J Urol 2018;36:1913-26. [Crossref] [PubMed]
7. O’Connor E, Timm B, Lawrentschuk N, et al. Open partial nephrectomy: current review. Transl Androl Urol 2020;9:3149-59.
8. Ljungberg B. Nephron-sparing surgery strategy: the current standard for the treatment of localised renal cell carcinoma. Eur Urol Suppl 2011;10:e49-e51.
9. Lee H, Lee M, Lee SE, et al. Outcomes of pathologic stage T3a renal cell carcinoma up-staged from small renal tumor: emphasis on partial nephrectomy. BMC Cancer 2018;18:427. [Crossref] [PubMed]
10. Liu Z, Cong X, Liu Z, et al. Adjuvant therapy for renal cell carcinoma: lessons from past failures and new opportunities in the era of immune checkpoint inhibition. Front Immunol 2026;17:1816253. [Crossref] [PubMed]
11. Motzer RJ, Tannir NM, McDermott DF, et al. Nivolumab plus Ipilimumab versus Sunitinib in Advanced Renal-Cell Carcinoma. N Engl J Med 2018;378:1277-90. [Crossref] [PubMed]
12. Rini BI, Plimack ER, Stus V, et al. Pembrolizumab plus Axitinib versus Sunitinib for Advanced Renal-Cell Carcinoma. N Engl J Med 2019;380:1116-27. [Crossref] [PubMed]
13. Braun DA, Bakouny Z, Hirsch L, et al. Beyond conventional immune-checkpoint inhibition - novel immunotherapies for renal cell carcinoma. Nat Rev Clin Oncol 2021;18:199-214. [Crossref] [PubMed]
14. Choueiri TK, Motzer RJ. Systemic Therapy for Metastatic Renal-Cell Carcinoma. N Engl J Med 2017;376:354-66. [Crossref] [PubMed]
15. Dankel SN, Svärd J, Matthä S, et al. COL6A3 expression in adipocytes associates with insulin resistance and depends on PPARγ and adipocyte size. Obesity (Silver Spring) 2014;22:1807-13. [Crossref] [PubMed]
16. Pasarica M, Gowronska-Kozak B, Burk D, et al. Adipose tissue collagen VI in obesity. J Clin Endocrinol Metab 2009;94:5155-62. [Crossref] [PubMed]
17. Khan T, Muise ES, Iyengar P, et al. Metabolic dysregulation and adipose tissue fibrosis: role of collagen VI. Mol Cell Biol 2009;29:1575-91. [Crossref] [PubMed]
18. Kanda H, Tateya S, Tamori Y, et al. MCP-1 contributes to macrophage infiltration into adipose tissue, insulin resistance, and hepatic steatosis in obesity. J Clin Invest 2006;116:1494-505. [Crossref] [PubMed]
19. Gesta S, Guntur K, Majumdar ID, et al. Reduced expression of collagen VI alpha 3 (COL6A3) confers resistance to inflammation-induced MCP1 expression in adipocytes. Obesity (Silver Spring) 2016;24:1695-703. [Crossref] [PubMed]
20. Guo HL, Chen G, Song ZL, et al. COL6A3 promotes cellular malignancy of osteosarcoma by activating the PI3K/AKT pathway. Rev Assoc Med Bras (1992) 2020;66:740-5. [Crossref] [PubMed]
21. Javali PS, Thirumurugan K. Artificial intelligence driven multi-omics framework identifies COL6A3 as a diagnostic biomarker and a putative gene target modulated by Embelin in Colorectal cancer. Front Oncol 2026;16:1711079. [Crossref] [PubMed]
22. Bankole NDA, Le Van T, Kerherve L, et al. Histo-Molecular Intratumoral Heterogeneity in Meningiomas: A Narrative Review. Cancers (Basel) 2026;18:1206. [Crossref] [PubMed]
23. Parihari S, Sarkar S, Vashishtha V, et al. Proteomics and Lipidomics Analysis Reveal That Membrane Remodeling and Extracellular Matrix Alterations Are Crucial for Cisplatin Resistance in Triple-Negative Breast Cancer. J Proteome Res 2026;25:2135-49. [Crossref] [PubMed]
24. Cao W, Zeng Z, Lan J, et al. Knockdown of FUT11 inhibits the progression of gastric cancer via the PI3K/AKT pathway. Heliyon 2023;9:e17600. [Crossref] [PubMed]
25. Xu C, Cao G, Zhang H, et al. Avellanin A Has an Antiproliferative Effect on TP-Induced RWPE-1 Cells via the PI3K-Akt Signalling Pathway. Mar Drugs 2024;22:275. [Crossref] [PubMed]
26. Wu Y, Xu Y. Integrated bioinformatics analysis of expression and gene regulation network of COL12A1 in colorectal cancer. Cancer Med 2020;9:4743-55. [Crossref] [PubMed]
27. Yan P, Wang Y, Meng X, et al. Whole Exome Sequencing of Ulcerative Colitis-associated Colorectal Cancer Based on Novel Somatic Mutations Identified in Chinese Patients. Inflamm Bowel Dis 2019;25:1293-301. [Crossref] [PubMed]
28. Webster BR, Rompre-Brodeur A, Daneshvar M, et al. Kidney cancer: from genes to therapy. Curr Probl Cancer 2021;45:100773. [Crossref] [PubMed]
29. U Gandhy S. Madan RA, Aragon-Ching JB. The immunotherapy revolution in genitourinary malignancies. Immunotherapy 2020;12:819-31. [Crossref] [PubMed]
30. Rowshanravan B, Halliday N, Sansom DM. CTLA-4: a moving target in immunotherapy. Blood 2018;131:58-67. [Crossref] [PubMed]
31. Rotte A. Combination of CTLA-4 and PD-1 blockers for treatment of cancer. J Exp Clin Cancer Res 2019;38:255. [Crossref] [PubMed]
32. Xu S, Liu D, Qin Z, et al. Experimental validation and pan-cancer analysis identified COL10A1 as a novel oncogene and potential therapeutic target in prostate cancer. Aging (Albany NY) 2023;15:15134-60. [Crossref] [PubMed]