Abstract
Background
Pancreatic adenocarcinoma (PAAD) is a highly lethal malignancy characterized by a dismal prognosis and limited treatment options. Protein tyrosine kinase 6 (PTK6) is abnormally activated in multiple cancer types. Understanding the function of PTK6 in pancreatic cancer could reveal novel therapeutic targets and prognostic markers.
Methods
We extracted the expression profiles and clinical data of patients with pancreatic adenocarcinoma from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO). The Wilcoxon rank-sum test was used to compare PTK6 gene expression levels between tumor tissues and normal pancreatic tissues. Functional enrichment analysis, Cox regression analysis, and Kaplan–Meier plot analysis were employed to determine the prognostic value of PTK6. Single-sample gene set enrichment analysis (ssGSEA) and TIMER3.0 were used to evaluate immune cell infiltration. Finally, CCK-8, RT‒qPCR, colony formation, and western blot assays were performed in vitro cellular experiments to verify the biological effects of PTK6 on the malignant behaviour of pancreatic cancer.
Results
PTK6 is highly expressed in pancreatic adenocarcinoma, with its upregulation validated in both TCGA (AUC = 0.935) and GEO (AUC = 0.796) cohorts. Elevated PTK6 expression was correlated with shorter overall survival (HR = 2.08, 95% CI: 1.36–3.19, p < 0.001), and multivariate analysis revealed that N1 stage was an independent prognostic factor for OS. In addition, PTK6 expression was negatively correlated with immune cell infiltration and key immune checkpoint genes, with the strongest correlation observed for TIGIT (ρ = -0.316, p < 0.001). In vitro experiments revealed that downregulating PTK6 expression via siRNA inhibited cell proliferation and colony formation, concomitant with increased LC3-II/LC3-I ratio, suggesting a potential role in autophagy regulation.
Conclusion
PTK6 could serve as a potential diagnostic and prognostic biomarker for PAAD and may participate in immune modulation and autophagy regulation. These findings, supported by multi-cohort bioinformatic analysis and preliminary experimental validation, highlight its prospective clinical value in prognosis evaluation, immunotherapy, and targeted therapy for PAAD.
Supplementary Information
The online version contains supplementary material available at 10.1007/s12672-026-05115-9.
Keywords: PTK6, Pancreatic adenocarcinoma, Immune infiltration, Prognosis, Autophagy, Biomarker
Introduction
Pancreatic cancer is a highly aggressive malignancy that often has nonspecific symptoms and a high mortality rate [1, 2]. Although surgical resection currently remains the primary effective curative treatment at an early stage, less than 20% of patients are eligible because of their advanced stage at diagnosis [3–5]. Treatment for advanced-stage patients primarily involves systemic chemotherapy combinations, such as FOLFIRINOX (a combination of 5-fluorouracil, leucovorin, irinotecan, and oxaliplatin) and gemcitabine plus nab-paclitaxel [6–8]. Over the past decade, improvements in diagnostic methods, perioperative management, radiotherapy techniques, and systemic treatments have contributed to prolonged survival; however, the 5-year survival rate of patients with pancreatic adenocarcinoma (PAAD) remains only 11%, making it one of the most challenging cancers to treat [9, 10].
Protein tyrosine kinase 6 (PTK6), also termed breast tumour kinase (Brk), plays a key role in regulating normal differentiation in a cell type- and context-dependent manner. In multiple cancer types, such as breast cancer, colorectal cancer and ovarian cancer [11, 12], PTK6 is often abnormally activated or mislocalized, implicating it in the altered signalling of malignant cells [13, 14]. Notably, PTK6 exhibits context-dependent functions—nuclear localization may exert tumor-suppressive effects through promoting differentiation, while cytoplasmic mislocalization drives oncogenic signaling [15]. This dual role, along with its established oncogenic functions in other epithelial cancers, makes PTK6 a compelling but poorly understood target in pancreatic cancer. Given the limited treatment options and poor prognosis of PAAD, exploration of the function of PTK6 is urgently needed as it may reveal novel therapeutic molecular targets and prognostic biomarkers in this disease.
Our study aimed to systematically evaluate the differential expression, prognostic value, immune cell infiltration and biological functions of PTK6 in PAAD using TCGA, GTEx and GEO databases. We analyzed correlations between PTK6 expression and clinicopathological parameters, clinical outcomes in various subgroups, and performed functional enrichment analysis in PAAD. Furthermore, in vitro experiments, including PTK6 knockdown, validated its involvement in regulating pancreatic cancer cell proliferation and autophagy.
In this study, PTK6 expression was significantly higher in PAAD tissues than in normal tissues. Increased PTK6 expression was significantly correlated with higher pathological stage, particularly in T3 compared with T2 and in stage II compared with stage I. Additionally, elevated PTK6 expression predicted poor outcomes, manifesting as the shorter overall survival (OS), progress-free interval (PFI) and disease specific survival (DSS). Multivariate Cox regression analyses revealed that stage N1 was an independent prognostic factor for OS in patients with PAAD. Furthermore, we constructed a reliable prognostic nomogram integrating TNM stage, age, and PTK6 expression to quantify the prognosis of PAAD patients. We observed that the prognosis of PAAD patients worsened as the total nomogram score increased. Single-sample gene set enrichment analysis (ssGSEA) revealed that PAAD patients with high PTK6 expression had substantially lower enrichment scores for TFH cells, cytotoxic cells, T cells, NK cells, DCs, and CD8 T cells, compared to those with low expression. Through immune infiltration analysis, we newly reveal that PTK6 expression is negatively correlated with key immune checkpoint genes, particularly TIGIT, suggesting a potential role in immune evasion. Furthermore, we provide the first evidence linking PTK6 to autophagy modulation in PAAD through GSEA, which identified significant enrichment of autophagy-regulatory pathways—including Rho GTPases and DNA damage response. Finally, protein-level validation using HPA database and In vitro experiments suggested a pro-tumorigenic role of PTK6.
Based on the findings of bioinformatics analyses and in vitro experiments, we concluded that PTK6 could serve as a significant prognostic biomarker and a crucial anticancer target in PAAD, while offering mechanistic insights into its regulation of autophagy and immune microenvironment.
Materials and methods
Data processing
In this study, miRNA expression profiles were obtained from the TCGA, GTEx and GEO databases. Xiantao Academic (https://www.xiantaozi.com/) was used for data processing and visualization. The platform’s modules for pancancer analysis, differential expression analysis, functional clustering, clinical significance evaluation, and subsequent statistical analyses were utilized. The cloud datasets form TCGA and GTEx used were composed of 171 normal tissue samples and 179 PAAD tumour tissue samples. Additionally, the microarray gene-expression dataset GSE62452 from GEO included 61 adjacent non-tumour tissue samples and 69 pancreatic tumor samples. The transcripts per million reads (TPM) method was used to remove technical biases in the sequenced data [16]. Data processing and visualization were transformed using log2 (value + 1) and conducted using R software (v4.2.1). These databases are publicly accessible.
Differential expression analysis of PTK6
The Wilcoxon rank-sum test was used to analyse the differential expression of PTK6 across cancers. High-expression and low-expression PTK6 groups were characterized statistically as having PTK6 expression levels above or below the median, respectively. The differential expression analysis was further evaluated with the ‘limma’ R package. The threshold for differentially expressed genes (DEGs) was defined as an adjusted p value < 0.05 and a |log FC| > 1. In addition, to assess the correlation of the top 20 DEGs with PTK6 expression, we performed the Spearman correlation analysis to measure the strength and direction of the monotonic associations between two variables [17]. Protein-level expression of PTK6 in pancreatic cancer was assessed using immunohistochemistry (IHC) data from the Human Protein Atlas (HPA) database (https://www.proteinatlas.org/). Two independent antibodies (HPA036070 and HPA036071) were used to validate PTK6 staining patterns in pancreatic cancer tissues and normal pancreatic tissues.
Clinical statistical analysis of PTK6 expression
Using the Wilcoxon rank-sum test, we investigated the relationships between the mRNA expression of PTK6 and clinicopathological characteristics, such as patient sex, patient age, patient histological grade, chronic pancreatitis in the patient, pathological stage, stage M, stage N, and stage T. A p value of less than 0.05 was considered to indicate statistical significance. The following significance symbols are used in this study: ns is equal to p ≥ 0.05; * is equal to p < 0.05, ** is equal to p < 0.01, *** is equal to p < 0.001, and **** is equal to p < 0.0001.
Kaplan‒Meier plotter analysis
The Kaplan‒Meier method was used to analyse overall survival (OS), disease-specific survival (DSS) and the progression-free interval (PFI). To assess the effect of clinical characteristics on patient prognosis and identify independent prognostic factors, we employed univariate and multivariate Cox regression analyses and constructed column charts. Additionally, we constructed nomogram plots with independent prognostic indicators and predicted survival at 1-, 2- and 3-year. A calibration chart was then used to verify the performance of the nomogram, thereby ensuring the reliability of our predictions. To assess the predictive performance of our prognostic model at specific time points, we used the timeROC package in R to calculate the time-dependent AUC values for 1-, 3-, and 5-year survival based on the risk score derived from our multivariate Cox model.
Functional enrichment analysis of PTK6 in PAAD
To identify the biological functions of PTK6 in PAAD, we conducted Gene Ontology (GO) analyses, Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses and gene set enrichment analysis (GSEA) using the “clusterProfiler” package in the R programming language. Patients with PAAD were divided into low and high-expression groups on the basis of the median expression level of PTK6. In addition, we performed GSEA using the c2.cp.all.v2022.1.Hs.symbols.gmt gene set collection (3,050 canonical pathways) from the Molecular Signatures Database (MSigDB). Pathways with an false discovery rate (FDR) of less than 0.25 and a p value of less than 0.05 were considered significantly enriched in the high PTK6 expression group.
Relationship between PTK6 expression and immune cell infiltration
Immune cell infiltration was assessed using two complementary algorithms. First, twenty-four different immune cells were examined in this study to determine the degree of immune infiltration. Their relative enrichment fraction in PAAD was measured utilizing ssGSEA, which was conducted using the R package “GSVA” [18]. To validate these results, we additionally employed the xCell algorithm via TIMER3.0 (http://timer.comp-genomics.org/). Furthermore, the correlation between PTK6 expression and immune cells was estimated using the Spearman correlation analysis. The Wilcoxon rank-sum test was then employed to determine the differences in immune cell infiltration levels between the high- and low-expression of PTK6 groups. To explore the potential immunomodulatory role of PTK6 in PAAD, we analysed the correlation between PTK6 expression and key immune checkpoint genes (PDCD1, CTLA4, CD274, LAG3, TIGIT). Heatmap visualization was generated using the platform’s “Heatmap” module with hierarchical clustering. Statistical significance was defined as p < 0.05.
Cell culture and siRNA transfection
Human pancreatic cancer cell lines, including Mia PaCa-2, SW1990, PANC1, BxPC-3, and Capan-1, were obtained from the American Type Culture Collection (ATCC, USA). These cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM), Iscove’s modified Dulbecco’s medium (IMDM) or RPMI 1640 medium (1640) supplemented with 10% foetal bovine serum (FBS), 100 µg/ml penicillin and 100 µg/ml streptomycin at 37 °C, 95% humidity, and 5% CO2. Small interfering RNA (siRNA) transfection was performed using the Lipofectamine™ 2000 transfection reagent following the manufacturer’s protocol. The sequences of the PTK6-targeting siRNAs were listed in Table 1.
Table 1.
The primer sequences for qPCR
| Gene | sense(5’−3’) | antisense(5’−3’) |
|---|---|---|
| siRNA 1 | GGUGGCCAUUAAGGUGAUUTT | AAUCACCUUAAUGGCCACCTT |
| siRNA 2 | GGAUUCUCCUGCAUGAGAUTT | AUCUCAUGCAGGAGAAUCCTT |
| siRNA 3 | GACCCCGUGUACAUCAUCATT | UGAUGAUGUACACGGGGUCTT |
| siRNA 4 | ACCUGGAGUCGCAGAAUUATT | UAAUUCUGCGACUCCAGGUTT |
RNA extraction and qRT‒PCR
Total RNA was extracted from cells using RNA isolation Total RNA Extraction Reagent (VAZYME, R401-01) according to the manufacturer’s instructions. Briefly, mRNA was reverse transcribed into cDNA using HiScript II Q RT SuperMix for qPCR. Real-time fluorescence quantitative PCR detection was performed with ChamQ SYBR qPCR Master Mix using a real-time PCR system (ABI, QuantStudio 6). The relative transcription level of PTK6 mRNA was subsequently calculated by the relative quantification method (2-ΔΔCT method), with GAPDH serving as the reference gene. The sequences of the qRT‒PCR primers used were as follows:
Homo-GAPDH:
Forward: 5′-ATGGGGAAGGTGAAGGTCGGAGT-3′
Reverse: 5′-TAGTTGAGGTCAATGAAGGGGTC-3′
Homo PTK6:
Forward: 5′- TATGTGGGCCTCTGGGACTT-3′
Reverse: 5′- GATGCAGCCAAAGAACCACG-3′
Western blot
Total protein in cells was extracted using RIPA lysis buffer supplemented with phenylmethanesulfonyl fluoride (PMSF; Meilunbio China) and centrifuged at 12,000 rpm at 4 °C for 5 min. A BCA assay kit (GBCBIO, China) was used to measure the total protein concentration. Equal amounts of denatured protein samples were subjected to 10% SDS‒PAGE. The protein sample was subsequently transferred to a 0.45 μm polyvinylidene difluoride (PVDF; Millipore, USA) membrane and blocked with 5% milk in Tris-buffered saline supplemented with 0.1% Tween-20 (TBST) at room temperature for 2 h. Next, the PVDF membrane was incubated with primary antibodies against GAPDH (1:10000; Proteintech) and PTK6 (1:1000; Proteintech) at 4 °C overnight. The next day, the membrane was washed with TBST three times for 5 min each and incubated with an HRP-conjugated anti-goat IgG-HRP secondary antibody at room temperature for 2 h.
CCK8 cell proliferation assay
Cell proliferation was measured using the Cell Counting Kit-8 (CCK-8; MCE, USA), and statistical analysis was performed. In brief, cell suspensions from the logarithmic growth phase at a concentration of 5 × 103 cells per 100 µl were seeded in 96-well plates and then incubated at 37 °C for 0, 24, 48, or 72 h. CCK-8 DMEM (10 µl of CCK-8 in 100 µl of DMEM) was added to each well at specific time points and incubated at 37 °C for an additional 2 h. The absorbance of each well at a wavelength of 450 nm (OD 450 nm) was then detected using a spectrophotometer.
Cell cloning
Cells were seeded at 200 cells per well in 6-well plates and cultured in complete medium supplemented with 10% FBS. After 10 days of culture, the colonies were subsequently fixed with 4% paraformaldehyde and stained with crystal violet. The photos were taken with a digital camera and analysed using ImageJ software.
Statistical analysis
Statistical analyses were performed using GraphPad Prism (version 10.6.0), and the results are presented as the means ± SDs. Multivariate or univariate Cox regression and logistic regression analyses were employed to analyse the correlations between PTK6 expression and clinicopathological variables. P values less than 0.05 were considered to indicate statistical significance.
Results
High expression of PTK6 in PAAD
Differential analysis of PTK6 expression in pan-cancer samples showed that the level of PTK6 was higher in most malignancies, such as pancreatic adenocarcinoma (PAAD), breast cancer (BRCA), and lung adenocarcinoma (LUAD), as shown in the boxplot in Fig. 1A. In unpaired samples from the TCGA-KIRC database, PTK6 expression was significantly higher in PAAD tissues than in normal tissues (p < 0.05). (Fig. 1B). Moreover, we conducted receiver operating characteristic (ROC) curve analysis, which is an effective method for evaluating the performance of diagnostic tests [19]. As shown in Fig. 1C, the area under the curve (AUC) of the ROC curve was 0.935 (95% CI: 0.904–0.967) at a cut-off of 2.241. To validate this finding, external validation was performed using GSE62452 dataset from the GPL6244 platform in the GEO database. Consistent with the TCGA results, PTK6 expression was also significantly higher in pancreatic tumor tissues than in adjacent non-tumor tissues (Fig. 1D). More importantly, ROC analysis yielded an AUC of 0.796 (95% CI: 0.719–0.873) at a cut-off of 4.608 (Fig. 1E). These results demonstrated that PTK6 could serve as a reproducible and a reliable biomarker for the diagnosis of PAAD.
Fig. 1.
The expression level of PTK6 across cancers and PAAD. A PTK6 expression in pancancer samples from the TCGA-KIRC database. B The expression of PTK6 in unpaired PAAD tissues compared with normal tissues in the TCGA database. C ROC curve for PTK6 as a diagnostic biomarker for PAAD (TCGA, AUC = 0.935). D Validation of PTK6 expression in adjacent non-tumor samples and PAAD samples using the GEO dataset (GSE62452). E ROC curve for PTK6 as a diagnostic biomarker for PAAD (GEO, AUC = 0.796)
Associations between PTK6 expression and clinicopathological factors and the value of PTK6 for prognosis and diagnosis in patients with PAAD
Clinical baseline characteristics (Table 2) were derived from PAAD samples in the TCGA database to assess the association between PTK6 expression and clinicopathological features. Percentages are calculated based on patients with available data. N/A indicates data missing and not available. As illustrated in Fig. 2A-D, increased PTK6 expression was significantly correlated with higher pathological stage, particularly in T3 compared with T2 (Fig. 2A, p < 0.05) and in stage II compared with stage I (Fig. 2D, p < 0.05). However, no significant association was found between pathologic stage N or M and PTK6 expression level (Fig. 2B and C). Additionally, compared with G1 tumours, histologic grade G2 and G3 tumours exhibited markedly elevated PTK6 expression (Fig. 2E, p < 0.01). As shown in Fig. 2F-H, high PTK6 expression was linked to chronic pancreatitis, sex and age (p < 0.05).
Table 2.
Clinicopathological characteristics associated with PTK6 expression in patients with PAAD from the TCGA database
| Characteristics | Low expression of PTK6 | High expression of PTK6 | P value |
|---|---|---|---|
| n | 89 | 89 | |
| Pathologic T stage, n (%) | 0.457 | ||
| T1&T2 | 18 (10.2%) | 13 (7.4%) | |
| T3 | 67 (38.1%) | 75 (42.6%) | |
| T4 | 2 (1.1%) | 1 (0.6%) | |
| N/A | 2 | 0 | |
| Pathologic N stage, n (%) | 0.647 | ||
| N0 | 23 (13.3%) | 26 (15%) | |
| N1 | 63 (36.4%) | 61 (35.3%) | |
| N/A | 3 | 2 | |
| Pathologic M stage, n (%) | 0.416 | ||
| M0 | 47 (56%) | 33 (39.3%) | |
| M1 | 1 (1.2%) | 3 (3.6%) | |
| N/A | 41 | 53 | |
| Histologic grade, n (%) | < 0.001 | ||
| G1 | 26 (14.8%) | 5 (2.8%) | |
| G2 | 43 (24.4%) | 52 (29.5%) | |
| G3 | 17 (9.7%) | 31 (17.6%) | |
| G4 | 2 (1.1%) | 0 (0%) | |
| N/A | 1 | 1 | |
| History of chronic pancreatitis, n (%) | 0.009 | ||
| No | 69 (48.6%) | 60 (42.3%) | |
| Yes | 2 (1.4%) | 11 (7.7%) | |
| N/A | 18 | 18 | |
| Gender, n (%) | 0.132 | ||
| Male | 44 (24.7%) | 54 (30.3%) | |
| Female | 45 (25.3%) | 35 (19.7%) | |
| Age, n (%) | 0.230 | ||
| <= 65 | 51 (28.7%) | 43 (24.2%) | |
| > 65 | 38 (21.3%) | 46 (25.8%) | |
| History of diabetes, n (%) | 0.368 | ||
| No | 51 (34.7%) | 58 (39.5%) | |
| Yes | 21 (14.3%) | 17 (11.6%) | |
| N/A | 17 | 14 | |
| Alcohol history, n (%) | 0.942 | ||
| No | 31 (18.7%) | 33 (19.9%) | |
| Yes | 50 (30.1%) | 52 (31.3%) | |
| N/A | 8 | 4 | |
| Smoker, n (%) | 0.575 | ||
| No | 34 (23.4%) | 32 (22.1%) | |
| Yes | 37 (25.5%) | 42 (29%) | |
| N/A | 18 | 15 | |
| Pathologic stage, n (%) | 0.442 | ||
| Stage I | 13 (7.4%) | 8 (4.6%) | |
| Stage II | 71 (40.6%) | 76 (43.4%) | |
| Stage III | 2 (1.1%) | 1 (0.6%) | |
| Stage IV | 1 (0.6%) | 3 (1.7%) | |
| N/A | 2 | 1 | |
| Race, n (%) | 0.245 | ||
| Asian | 5 (2.9%) | 6 (3.4%) | |
| White | 77 (44.3%) | 80 (46%) | |
| Black or African American | 5 (2.9%) | 1 (0.6%) | |
| N/A | 2 | 2 |
Data are shown as n (%). Missing data are marked as “N/A”. Percentages are based on available cases
Fig. 2.
Correlation of PTK6 expression with clinicopathological parameters and survival in PAAD. (A-H) PTK6 expression in different subgroups: A T stage, B stage N, C stage M, D pathological stage, E histologic grade, F chronic pancreatitis, G gender, H age. I-K Kaplan‒Meier curves for OS (I), PFS (J) and DSS (K). Statistical significance was assessed using a Wilcoxon rank-sum test or Kruskal-Wallis test, and log-rank test. *p < 0.05, **p < 0.01, ***p < 0.001
Kaplan‒Meier (KM) analysis was employed to examine the prognostic significance of PTK6 in PAAD. Patients were classified by PTK6 expression levels using the median value as the cut-off. Thus, patients were subsequently grouped into two categories: the high PTK6 expression and low PTK6 expression groups. Patients with high PTK6 expression had worse OS than those with low PTK6 expression (HR = 2.08, 95% CI: 1.36–3.19, p < 0.001) (Fig. 2I). Additionally, the PFI and DSS tended to be shorter in the high PTK6 expression group (Fig. 2J and K).
PTK6 nomogram construction and verification
To gain deeper insights into prognostic factors in patients with PAAD, we subsequently performed both univariate and multivariate Cox regression analyses. Univariate analysis of OS revealed that PTK6 expression, stage T3, stage N1, grade G2 and G3 were significantly different (Fig. 3A, p < 0.05). Additionally, the multivariate analysis revealed that stage N1 was an independent prognostic factor for OS in patients with PAAD (Fig. 3B, p = 0.01). Furthermore, on the basis of the Cox regression analysis, we constructed a prognostic nomogram using TNM stage, grade, age, gender, somking, diabetes, alcohol consumption and PTK6 expression to quantify the prognosis of PAAD patients. As shown in Fig. 3C, we observed that the prognosis of PAAD patients decreased as the total number of points increased. Afterwards, we produced calibration curves, as shown in Fig. 3D, to test the accuracy and reliability of the predictive performance of the nomogram. The calibration curve lay close to the 45° ideal line, demonstrating good accuracy in predicting actual survival probabilities. Additionally, the C-index was 0.676 (95% CI: 0.626–0.726), indicating moderate discriminative ability. The time-dependent AUCs for 1‑, 3‑, and 5‑year survival were 0.659 (95% CI: 0.562–0.755), 0.718 (95% CI: 0.581–0.854), and 0.735 (95% CI: 0.505–0.963), respectively, indicating moderate to good predictive accuracy of the prognostic model (Fig. 3E).
Fig. 3.
Construction and verification of the PTK6 prognostic model. Forest map of OS in patients with PAAD based on A univariate and B multivariate Cox regression analyses. C A nomogram for predicting the OS rates of patients with PAAD at 1, 2 and 3 years. D Calibration curves for the 1-, 2- and 3-year OS survival nomograms. E The AUC values for 1-, 2-, and 3-year OS were 0.659 (95% CI: 0.562–0.755), 0.718 (95% CI: 0.581–0.854), and 0.735 (95% CI: 0.505–0.963), respectively
Analysis of functional enrichment of differentially expressed genes (DEGs) linked to PTK6
Following the single-gene differential expression analysis of PTK6, we screened 1845 genes that met the predefined criteria. Among these genes, 443 (24.0%) were significantly upregulated (adjusted p value < 0.05 and | log2 FC |>1.5), whereas 1402 (76%) were downregulated (Fig. 4A). The study further concentrated on the top 10 upregulated and top 10 downregulated DEGs ranked by log2 fold change (e.g., CASP14, REG4, MUC21, CYMP-AS1, SCGB2A2) and examined their correlation with PTK6 expression (Fig. 4B). Functional enrichment analyses, including GO terms and KEGG pathways, were then performed for all the DEGs. The results indicated that biological process (BP) terms were enriched primarily in keratinization, epidermal differentiation, and intermediate filament organization, pathways associated with EMT and cell motility. The molecular function (MF) category was enriched predominantly for peptidase inhibitor activity and structural constituent of cytoskeleton, linking PTK6 to ECM remodelling and cytoskeletal dynamics. At the cellular component (CC) level, enrichment was observed in keratin filament and intermediate filament cytoskeleton, consistent with PTK6’s role in cytoskeletal organization [20]. With respect to the results of the KEGG pathway analysis, significant enrichment was observed in pancreatic secretion, fat digestion and absorption, protein digestion and absorption, reflecting the tissue-specific origin of pancreatic cancer (Fig. 4C and Supplementary Table 1). To further explore the functional implications of PTK6 in pancreatic cancers, we performed GSEA. As shown in Fig. 4D, the analysis revealed that multiple pathways were significantly enriched in the high PTK6 expression group, including Rho GTPases activate PKNs, DNA methylation, condensation of prophase chromosomes, keratinization, recognition and association of DNA glycosylase with site containing an affected purine, and formation of cornified envelope (adjusted p value < 0.05 and FDR < 0.25).
Fig. 4.
PTK6-associated DEGs and functional enrichment analyses through GSEA, GO, and KEGG. A Volcano plot visualization of significantly downregulated (blue) and upregulated (red) DEGs. B Heatmap showing the top 10 upregulated and top 10 downregulated DEGs (ranked by log2FC) in PAAD versus normal tissues. Expression values are log2(TPM + 1) transformed. C GO enrichment analysis of PTK6-correlated genes (BP, CC, MF). D GSEA plots showing significantly enriched pathways in PTK6-high tumors, including Rho GTPases activate PKNs and recognition and association of DNA glycosylase with site containing an affected purine
Correlations between PTK6 levels and immune cell infiltration
Our study assessed the correlation between PTK6 levels and 24 immune cell subsets. The results revealed significant differences between PTK6 expression levels and immune cell enrichment across cancers, as shown in the heatmap in Fig. 5A. Importantly, as shown in Fig. 5B, the results obtained with xCell were highly consistent with our ssGSEA-based findings. The heatmap clearly demonstrates that PTK6 is most strongly negatively correlated with CD8 T cells, M2 macrophages, and NK cells, myeloid dendritic cells (p < 0.05). We noticed that it negatively correlated with most immune infiltrates in patients with PAAD (Fig. 5C). Notably, strong negative correlations were observed for TFH cells and cytotoxic cells (Fig. 5D and E; |R|≥ 0.3). Comparable trends were also evident for T cells, NK cells, DC, and CD8 T cells (Fig. 5F-I). We further systematically examined 24 immune cell types stratified by PTK6 expression levels in patients with PAAD (Fig. 6A). Comparative analysis revealed that compared with patients with low PTK6 expression, patients with PAAD with high PTK6 expression had substantially lower enrichment scores for TFH and cytotoxic cells, T cells, NK cells, DCs, and CD8 T cells, as illustrated in the boxplot in Fig. 6B-G. Spearman correlation analysis revealed that PTK6 expression was negatively correlated with key immune checkpoint genes, including PDCD1, CTLA4, CD274, LAG3, and TIGIT (Fig. 7A-E). The strongest negative correlation was observed with TIGIT (Spearman’s R = − 0.316, p < 0.001). A heatmap further illustrated the inverse expression patterns of PTK6 and these immune checkpoint genes across individual samples (Fig. 7F).
Fig. 5.
Correlation between PTK6 expression level and immune cell infiltration. A Pan-cancer heatmap of the correlation between PTK6 expression levels and immune cell enrichment (TCGA). B Validation using xCell algorithm (TIMER3.0). C Lollipop plot representing the correlation between PTK6 expression levels and the infiltration of 24 immune cell types in patients with PAAD based on ssGSEA analysis. D-I Scatter plots for D TFH, E cytotoxic cell, F T cell, G NK cell, H DC, and I CD8 T-cells, with regression lines and 95% confidence intervals. *p < 0.05, **p < 0.01, ***p < 0.001
Fig. 6.
Immune cell enrichment analysis based on the expression of PTK6. A Boxplots illustrated the enrichment scores of 24 immune cell types in patients with PAAD with high and low PTK6 expression. B-G Boxplots comparing the infiltration levels of B TFH cells, C cytotoxic cells, D T cells, E NK cells, F DC cells, and G CD8 T cells. Two-tailed unpaired t test for comparisons, *p < 0.05, **p < 0.01, ***p < 0.001
Fig. 7.
PTK6 expression negatively correlates with immune checkpoint genes in PAAD. A-E Scatter plots showing the negative correlations between PTK6 expression and A CD274, B LAG3, C PDCD1, D CTLA4, and E TIGIT in the TCGA cohort, with Spearman’s R and p values. F Heatmap showing the expression patterns of PTK6 and the five immune checkpoint genes across PAAD samples. *p < 0.05, **p < 0.01, ***p < 0.001
PTK6 promotes proliferation and inhibits autophagy in pancreatic cancer
Based on the GSEA results, several of these enriched pathways particularly Rho GTPases activate PKNs and recognition and association of DNA glycosylase with site containing an affected purine have been functionally linked to autophagy regulation [21, 22]. Recent bioinformatics studies identified PTK6 as an autophagy-related exosome gene in pancreatic cancer [23]. However, the specific mechanisms by which PTK6 regulates autophagy remain unexplored. To validate our transcriptomic findings at the protein level, we first examined PTK6 expression using immunohistochemistry (IHC) data from the Human Protein Atlas (HPA) database. As shown in Fig. 8A, PTK6 exhibited moderate to strong cytoplasmic positivity in PAAD tissues, whereas normal pancreatic tissues showed weak staining. This protein-level observation was consistent with our TCGA and GEO mRNA expression analyses. To further explore the potential role of PTK6 in pancreatic cancer, we conducted cell function experiments. We found that the mRNA and protein levels of PTK6 in pancreatic cancer cell lines, specifically Mia PaCa-2 cells, were significantly greater in Mia PaCa-2 cells than in normal human pancreatic duct epithelial cells (HPDE) (Fig. 8B). To further evaluate the oncogenic functions of PTK6 in pancreatic cancer, we transfected Mia PaCa-2 cells with siRNA and confirmed the knockdown efficiency at both the mRNA and protein levels (Fig. 8C). Consequently, PTK6 downregulation resulted in significant suppression of cell proliferation (p < 0.05; Fig. 8D) and a marked reduction in colony-forming capacity (Fig. 8E). Quantification of colony formation was shown in Fig. 8F. To validate the role of PTK6 in modulating autophagy, we compared differences in autophagic protein expression between the silencing PTK6 group and the control group. As shown in Fig. 8G, silencing PTK6 in Mia PaCa-2 cells increased the LC3-II/LC3-I ratio, suggesting a potential involvement of PTK6 in the regulation of autophagy.
Fig. 8.
PTK6 expression in PAAD tissues and its effects on proliferation, colony formation, and autophagy. A IHC staining of PTK6 in normal pancreas and PAAD tissues (HPA database). Scale bar: 200 μm. B mRNA and protein levels of PTK6 in pancreatic cancer cell lines and HPDE cells. C Verification of the mRNA and protein levels of si-PTK6 in Mia PaCa-2 cells. D Cell proliferation assessed by CCK-8 assay in Mia PaCa-2 cells after PTK6 knockdown. E Colony formation assay. F Quantification of colony formation. G LC3-I, LC3‐II, and GAPDH protein expression levels were detected by western blot analysis. Data are shown as mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001
Discussion
Protein tyrosine kinase 6 (PTK6), also known as BRK or Sik, is an SRC-related intracellular tyrosine kinase that is predominantly located in epithelial cells [24, 25]. PTK6 is defined by its characteristic triad of structural domains, SRC homology 3 (SH3), SRC homology 2 (SH2) and a tyrosine kinase catalytic domain (TyrKc), which allow the phosphorylation of various substrate proteins involved in signal transduction processes [11, 12, 26]. Unlike SRC family members, PTK6 lacks the SH4 domain required for lipid modification and membrane association, thereby facilitating flexibility in its intracellular localization [27].
In this study, we provide the first evidence linking PTK6 to both immune modulation and autophagy regulation in pancreatic cancer, forming a potential mechanistic framework. Our study revealed that PTK6 is highly expressed in PAAD and is significantly correlated with a shorter overall survival. Elevated PTK6 expression was found to be associated with chronic pancreatitis, sex, and age. Multivariate analysis revealed that stage N1 was an independent prognostic factor for OS in patients with PAAD. In addition, high PTK6 expression in patients with PAAD is negatively correlated with immune cell subsets, including TFH cells, cytotoxic cells, T cells, NK cells, DCs, and CD8 T cells, and with key immune checkpoint genes, particularly TIGIT. These results suggest that PTK6-high of PAAD may establish an immunosuppressive microenvironment by reducing immune cell infiltration and suppressing immune checkpoint expression, potentially serving as a predictor of immune status.
Furthermore, GSEA revealed significant enrichment of several pathways that have been functionally linked to autophagy regulation, including Rho GTPases activate PKNs and recognition of DNA glycosylase with site containing an affected purine. Rho GTPases (RAC1, RHOA, CDC42) are well-established regulators of autophagosome formation and maturation, and recent bioinformatics analysis revealed that PTK6 is potentially a differentially expressed autophagy-related gene (ARG) [28, 29], and DNA damage response pathways are tightly coupled with autophagy activation [30]. Chen et al. [11] reported that PTK6 physically interacted with HNRNPH1 and triggered splicing-switching of the NBR1 exon 10 inclusion transcript, thereby activating autophagy and inhibiting apoptosis in CRC, identifying a novel PTK6-HNRNPH1-NBR1 regulatory autophagy axis. Consistent with these findings, our in vitro experiments demonstrated that PTK6 downregulation via siRNA inhibited proliferation and colony formation and enhanced autophagy, suggesting that PTK6 may regulate autophagy and immune modulation in pancreatic cancer.
PTK6 has been studied in various cancers, including breast, prostate, and colorectal cancer [31, 32]. In prostate cancer cells, PTK6 overexpression stimulates the epithelial‒mesenchymal transition (EMT) by activating the AKT pathway and promoting cancer cell migration and metastasis in xenograft models [27, 31]. In breast cancer, PTK6 enhances EGF-induced cell migration, whereas the downregulation of PTK6 expression via shRNA or the specific proteolysis-targeting compound MS105 effectively induces apoptosis and inhibits the growth and migration of breast cancer cells, which is consistent with observations in colorectal cancer [33, 34]. However, conflicting findings have been reported. For example, Ma et al. [35]. reported that PTK6 overexpression suppresses the migration of oesophageal cancer cells. The mechanism involves a PTK6-induced reduction in the phosphorylation of Akt and GSK3β, leading to β-catenin activation. Our study extends these findings by demonstrating that PTK6 is highly expressed in PAAD and significantly correlated with poor prognosis, positioning PTK6 as a promising biomarker candidate. Its links to immune infiltration and autophagy also suggest possible therapeutic relevance.
However, this study has several limitations. First, although we incorporated ssGSEA and xCell algorithms to estimate immune infiltration, which provides concordant evidence for PTK6’s potential role in immune cell infiltration in PAAD, further molecules and functional studies are required. Second, our experimental work was preliminary and lacked comprehensive functional assays across multiple cell lines. The connection between PTK6 and autophagy also requires more direct validation using established flux assays.
Conclusions
In conclusion, our study identifies PTK6 as a potential prognostic biomarker and therapeutic target in pancreatic cancer, with links to autophagy and immune regulation. However, the underlying molecular mechanisms require further validation.
Supplementary Information
Acknowledgements
We sincerely thank the TCGA, HPA and GEO databases for providing valuable data, and the R software for supporting our statistical analysis. We also thank all researchers and participants for their contributions.
Accordance statement
This study was conducted in accordance with the relevant guidelines and regulations.
Author contributions
The study was designed by Y. Fan, F. Hou and F. Wang; materials and methods were contributed by F. Wang, S. Yu, P. Yang and J. Jiao; and F. Wang and S. Yu were responsible for performing the experiments and analysing the data. F. Wang were in charge of editing the draft. All the authors reviewed the final manuscript.
Funding
This work was supported by the National Natural Science Foundation of China (grant no. 82405104 to Feng-jiao Wang and grant no. 82405116 to Shu-lin Yu); National Science and Technology Major Project (grant. no. 2025ZD0545000、2025ZD0545002 to Yue Fan).
Data availability
The datasets generated and analysed in this study are available directly via the following web links: the TCGA database (https://port al.gdc.cancer.gov/), Kaplan–Meier Plotter ([http://kmplot.com/analysis/](http:/kmplot.com/analysis)), GEO database ([https://www.ncbi.nlm.nih.gov/geo/](https:/www.ncbi.nlm.nih.gov/geo)), TIMER3.0 ([http://timer.comp-genomics.org/](http:/timer.comp-genomics.org)), HPA ([https://www.proteinatlas.org/](https:/www.proteinatlas.org)) and Xiantao Academic (https://www.xiantaozi.com/).
Declarations
Ethics approval and consent to participate
This study was based on publicly available data from TCGA, HPA and GEO databases. No ethical approval was required. All methods were conducted in accordance with the relevant guidelines and regulations.
Consent for publication
All the authors have approved to the publication of the final manuscript.
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.
Feng-jiao Wang and Shu-lin Yu have contributed equally to this work.
References
- 1.Stoop TF, Javed AA, Oba A, Koerkamp BG, Seufferlein T, Wilmink JW, Besselink MG. Pancreatic cancer. Lancet. 2025;405(10485):1182–202. [DOI] [PubMed] [Google Scholar]
- 2.Huang J, Lok V, Ngai CH, Zhang L, Yuan J, Lao XQ, Ng K, Chong C, Zheng ZJ, Wong MCS. Worldwide burden of, risk factors for, and trends in pancreatic cancer. Gastroenterology. 2021;160(3):744–54. [DOI] [PubMed] [Google Scholar]
- 3.Huang L, Jansen L, Balavarca Y, Molina-Montes E, Babaei M, van der Geest L, Lemmens V, Van Eycken L, De Schutter H, Johannesen TB, et al. Resection of pancreatic cancer in Europe and USA: An international large-scale study highlighting large variations. Gut. 2019;68(1):130–9. [DOI] [PubMed] [Google Scholar]
- 4.Pedrazzoli S. Surgical treatment of pancreatic cancer: Currently debated topics on morbidity, mortality, and lymphadenectomy. Surg Oncol. 2022;45:101858. [DOI] [PubMed] [Google Scholar]
- 5.Strobel O, Neoptolemos J, Jäger D, Büchler MW. Optimizing the outcomes of pancreatic cancer surgery. Nat Rev Clin Oncol. 2019;16(1):11–26. [DOI] [PubMed] [Google Scholar]
- 6.Brahmer JR, Lacchetti C, Schneider BJ, Atkins MB, Brassil KJ, Caterino JM, Chau I, Ernstoff MS, Gardner JM, Ginex P, et al. Management of immune-related adverse events in patients treated with immune checkpoint inhibitor therapy: American society of clinical oncology clinical practice guideline. J Clin Oncol. 2018;36(17):1714–68. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Ghaneh P, Palmer D, Cicconi S, Jackson R, Halloran CM, Rawcliffe C, Sripadam R, Mukherjee S, Soonawalla Z, Wadsley J, et al. Immediate surgery compared with short-course neoadjuvant gemcitabine plus capecitabine, FOLFIRINOX, or chemoradiotherapy in patients with borderline resectable pancreatic cancer (ESPAC5): A four-arm, multicentre, randomised, phase 2 trial. Lancet Gastroenterol Hepatol. 2023;8(2):157–68. [DOI] [PubMed] [Google Scholar]
- 8.Kunzmann V, Siveke JT, Algül H, Goekkurt E, Siegler G, Martens U, Waldschmidt D, Pelzer U, Fuchs M, Kullmann F, et al. Nab-paclitaxel plus gemcitabine versus nab-paclitaxel plus gemcitabine followed by FOLFIRINOX induction chemotherapy in locally advanced pancreatic cancer (NEOLAP-AIO-PAK-0113): A multicentre, randomised, phase 2 trial. Lancet Gastroenterol Hepatol. 2021;6(2):128–38. [DOI] [PubMed] [Google Scholar]
- 9.Mizrahi JD, Surana R, Valle JW, Shroff RT. Pancreatic cancer. Lancet. 2020;395(10242):2008–20. [DOI] [PubMed] [Google Scholar]
- 10.Stoffel EM, Brand RE, Goggins M. Pancreatic cancer: Changing epidemiology and new approaches to risk assessment, early detection, and prevention. Gastroenterology. 2023;164(5):752–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Chen B, Liu B, Chen J, Li W, Ma N, Liu J, Fan R, Hu Q, Song H, Xu Y, et al. PTK6 drives HNRNPH1 phase separation to activate autophagy and suppress apoptosis in colorectal cancer. Autophagy. 2025;21(8):1680–99. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Gilic MB, Tyner AL. Targeting protein tyrosine kinase 6 in cancer. Biochim Biophys Acta Rev Cancer. 2020;1874(2):188432. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Ostrander JH, Daniel AR, Lange CA. Brk/PTK6 signaling in normal and cancer cell models. Curr Opin Pharmacol. 2010;10(6):662–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Lang YD, Chen HY, Ho CM, Shih JH, Hsu EC, Shen R, Lee YC, Chen JW, Wu CY, Yeh HW, et al. PSPC1-interchanged interactions with PTK6 and β-catenin synergize oncogenic subcellular translocations and tumor progression. Nat Commun. 2019;10(1):5716. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Haegebarth A, Bie W, Yang R, Crawford SE, Vasioukhin V, Fuchs E, Tyner AL. Protein tyrosine kinase 6 negatively regulates growth and promotes enterocyte differentiation in the small intestine. Mol Cell Biol. 2006;26(13):4949–57. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Zhao S, Ye Z, Stanton R. Misuse of RPKM or TPM normalization when comparing across samples and sequencing protocols. RNA. 2020;26(8):903–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Sedgwick P. Spearman’s rank correlation coefficient. BMJ. 2014;349:g7327. [DOI] [PubMed] [Google Scholar]
- 18.Bindea G, Mlecnik B, Tosolini M, Kirilovsky A, Waldner M, Obenauf AC, Angell H, Fredriksen T, Lafontaine L, Berger A, et al. Spatiotemporal dynamics of intratumoral immune cells reveal the immune landscape in human cancer. Immunity. 2013;39(4):782–95. [DOI] [PubMed] [Google Scholar]
- 19.Mandrekar JN. Receiver operating characteristic curve in diagnostic test assessment. J Thorac Oncol. 2010;5(9):1315–6. [DOI] [PubMed] [Google Scholar]
- 20.Alwanian WM, Vlajic K, Bie W, Kajdacsy-Balla A, Tyner AL. Protein tyrosine kinase 6 regulates activation of SRC kinase. J Biol Chem. 2022;298(11):102584. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Huo L, Huang X, Wang Y, Ouyang Y, Zheng X, Ouyang Y, Cao X, Chen K, Wei D, Wu Y, et al. RAB33A promotes metastasis via RhoC accumulation through non-canonical autophagy in cervical cancer. Cell Death Dis. 2025;16(1):130. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Ye Y, Lin P, Zhang W, Tan S, Zhou X, Li R, Pu Q, Koff JL, Dhasarathy A, Ma F, et al. DNA Repair Interacts with Autophagy To Regulate Inflammatory Responses to Pulmonary Hyperoxia. J Immunol. 2017;198(7):2844–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Liang L, Guo P, Xiao M, Zu F, Hao G, Zhang C, Liu Q. Prognostic analysis and mechanistic exploration of the autophagy-related exosome genes ITGA3, ITGB4, and PTK6 in pancreatic ductal adenocarcinoma. Discov Oncol. 2025;16(1):1204. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Box C, Pennington C, Hare S, Porter S, Edwards D, Eccles S, Crompton M, Harvey A. Brk/PTK6 and involucrin expression may predict breast cancer cell responses to vitamin D3. Int J Mol Sci. 2023;24(13):10757. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Anderson TMR, Ma SH, Raj GV, Cidlowski JA, Helle TM, Knutson TP, Krutilina RI, Seagroves TN, Lange CA. Breast tumor kinase (Brk/PTK6) is induced by HIF, glucocorticoid receptor, and PELP1-mediated stress signaling in triple-negative breast cancer. Cancer Res. 2016;76(6):1653–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Ang HL, Yuan Y, Lai X, Tan TZ, Wang L, Huang BB, Pandey V, Huang RY, Lobie PE, Goh BC, et al. Putting the BRK on breast cancer: From molecular target to therapeutics. Theranostics. 2021;11(3):1115–28. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Zheng Y, Wang Z, Bie W, Brauer PM, White BEP, Li J, Nogueira V, Raychaudhuri P, Hay N, Tonetti DA, et al. PTK6 activation at the membrane regulates epithelial-mesenchymal transition in prostate cancer. Cancer Res. 2013;73(17):5426–37. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Aguilera MO, Berón W, Colombo MI. The actin cytoskeleton participates in the early events of autophagosome formation upon starvation induced autophagy. Autophagy. 2012;8(11):1590–603. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Zhou Q, Chen X, Chen Q, Liu H, Hao L. A four autophagy-related gene-based prognostic signature for pancreatic cancer. Crit Rev Eukaryot Gene Expr. 2021;31(4):89–100. [DOI] [PubMed] [Google Scholar]
- 30.Xia L, Xing Y, Ye X, Wu Y, Yang Y, Yin Z, Wang A, Chen J, Zhang M. TRIM21-driven K63-linked ubiquitination of RBM38c, as a novel interactor of BECN1, contributes to DNA damage-induced autophagy. Cell Death Differ. 2025;32(7):1317–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Alwanian WM, Tyner AL. Protein tyrosine kinase 6 signaling in prostate cancer. Am J Clin Exp Urol. 2020;8(1):1–8. [PMC free article] [PubMed] [Google Scholar]
- 32.Sahu R, Pattanayak SP. Strategic developments & future perspective on gene therapy for breast cancer: Role of mTOR and Brk/ PTK6 as molecular targets. Curr Gene Ther. 2020;20(4):237–58. [DOI] [PubMed] [Google Scholar]
- 33.Mathur PS, Gierut JJ, Guzman G, Xie H, Xicola RM, Llor X, Chastkofsky MI, Perekatt AO, Tyner AL. Kinase-dependent and -independent roles for PTK6 in colon cancer. Mol Cancer Res. 2016;14(6):563–73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Martinez C, Xiong Y, Bartkowski A, Harada I, Ren X, Byerly J, Port E, Jin J, Irie H. A PROTAC degrader suppresses oncogenic functions of PTK6, inducing apoptosis of breast cancer cells. Cell Chem Biol. 2025;32(2):255–e266258. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Ma S, Bao JYJ, Kwan PS, Chan YP, Tong CM, Fu L, Zhang N, Tong AHY, Qin YR, Tsao SW, et al. Identification of PTK6, via RNA sequencing analysis, as a suppressor of esophageal squamous cell carcinoma. Gastroenterology. 2012;143(3):675–e686612. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets generated and analysed in this study are available directly via the following web links: the TCGA database (https://port al.gdc.cancer.gov/), Kaplan–Meier Plotter ([http://kmplot.com/analysis/](http:/kmplot.com/analysis)), GEO database ([https://www.ncbi.nlm.nih.gov/geo/](https:/www.ncbi.nlm.nih.gov/geo)), TIMER3.0 ([http://timer.comp-genomics.org/](http:/timer.comp-genomics.org)), HPA ([https://www.proteinatlas.org/](https:/www.proteinatlas.org)) and Xiantao Academic (https://www.xiantaozi.com/).