Prognostic and Immunological Significance of TIPARP in Pancreatic Cancer

Abstract

Background

Pancreatic ductal adenocarcinoma (PAAD) remains one of the most lethal malignancies, with a 5-year survival rate below 10%. TCDD-inducible poly (ADP-ribose) polymerase (TIPARP) is aberrantly expressed in various tumors, but its mechanistic role and clinical significance in pancreatic cancer remain unclear.

Methods

We analyzed TIPARP expression patterns and prognostic relevance by integrating data from the TCGA-PAAD (n = 178) and GTEx (n = 171) databases. Single-cell RNA sequencing, immune infiltration profiling, and Mendelian randomization analyses were conducted to assess cell-type specificity and causal relationships. The biological functions of TIPARP were validated through in vitro assays, and downstream mechanisms were explored via pathway enrichment analysis.

Results

TIPARP was significantly upregulated in pancreatic cancer tissues and correlated with worse overall survival (P = 5.9e − 3) and progression-free survival (P = 0.01), serving as an independent prognostic risk factor. A TIPARP-based nomogram exhibited strong predictive performance (C-index = 0.666; 5-year AUC = 0.885). Single-cell transcriptomics identified fibroblasts as the primary TIPARP-expressing cell type. Immune profiling revealed that high TIPARP expression was associated with altered immune cell infiltration and positively correlated with immune checkpoint molecules such as PD-L1 and TIM-3. Mendelian randomization confirmed a causal association between genetically predicted TIPARP expression and increased PAAD risk (OR = 1.511, P = 0.008). Functional assays showed that TIPARP promotes cell proliferation, migration, and invasion, while pathway analysis implicated the PI3K-Akt and ECM-receptor interaction signaling cascades.

Conclusion

TIPARP is a novel independent prognostic biomarker and potential therapeutic target in pancreatic cancer. It contributes to tumor progression by promoting an immunosuppressive microenvironment and activating pro-tumorigenic signaling. These findings highlight TIPARP as a promising candidate for precision medicine and combination immunotherapy in PAAD.

Introduction

Pancreatic ductal adenocarcinoma (PAAD), often referred to as the “king of cancers,” is the fourth leading cause of cancer-related mortality worldwide and is projected to rise to second place by 2030 [1]. Characterized by an insidious onset, aggressive biology, early metastasis, and resistance to conventional therapies, PAAD has a dismal 5-year survival rate of approximately 10%—significantly lower than that of most other solid tumors [2]. Despite advances in comprehensive treatment strategies, including surgery, chemotherapy, and radiotherapy, the overall prognosis of pancreatic cancer has shown little improvement over the past decades [3, 4]. Therefore, a deeper understanding of the molecular pathogenesis of PAAD and the identification of novel prognostic biomarkers and therapeutic targets are urgently needed to improve clinical outcomes [5, 6].

TCDD-inducible poly(ADP-ribose) polymerase (TIPARP), also known as PARP7, is a unique member of the PARP family [7]. Unlike the canonical PARP1 and PARP2, TIPARP exhibits mono-ADP-ribosyltransferase activity and participates in diverse biological processes, including transcriptional regulation, DNA damage response, inflammation, and immune modulation [8, 9]. Recent studies have highlighted TIPARP’s role in regulating cellular stress responses by ADP-ribosylating key transcription factors and signaling molecules [10, 11]. TIPARP is increasingly recognized in the context of tumor biology, with aberrant expression reported in hepatocellular carcinoma, breast cancer, and colorectal cancer, where it is implicated in tumor proliferation, invasion, metastasis, and therapeutic resistance [12, 13]. Given its immunomodulatory potential—and considering the uniquely immunosuppressive tumor microenvironment (TME) of pancreatic cancer—investigating the functional role of TIPARP in this context is of particular relevance [14].

The TME of PAAD is defined by several hallmark features, including dense desmoplastic stroma, hypovascularity, and profound immunosuppression [15, 16]. This specialized environment not only supports tumor cell survival and proliferation but also poses a significant barrier to effective drug delivery and immunotherapy [17, 18]. It is enriched with cancer-associated fibroblasts (CAFs), immunosuppressive macrophages, and regulatory T cells, whereas effector CD8⁺ T cell function is severely compromised [19]. Moreover, immune checkpoint molecules such as PD-1/PD-L1, CTLA-4, and TIM-3 are highly expressed, establishing a complex immune evasion network that underlies the poor efficacy of current immunotherapies [15, 20]. Overcoming this immunosuppressive barrier remains a major therapeutic challenge in PAAD.

Although the role of TIPARP in other malignancies has been preliminarily explored [12, 13], its expression landscape, clinical significance (particularly its prognostic value), underlying molecular mechanisms, and potential involvement in shaping the immunosuppressive TME of PAAD remain poorly understood. Specifically, key questions remain unanswered: Does TIPARP contribute to immune modulation in the PAAD microenvironment? Can its expression predict patient prognosis or response to immunotherapy? Is TIPARP a viable therapeutic target in pancreatic cancer? Additionally, traditional observational studies cannot establish causality between TIPARP expression and tumor progression [21], limiting its utility as a rational therapeutic target.

To address these critical questions, we performed a comprehensive multi-dimensional analysis integrating large-scale public datasets, single-cell transcriptomics, Mendelian randomization (MR), and in vitro functional assays to systematically investigate the biological function and clinical relevance of TIPARP in pancreatic cancer. Our findings not only provide new insights into the molecular underpinnings of PAAD but also propose TIPARP as a potential biomarker and therapeutic target for precision oncology and immunotherapy in this devastating disease.

Materials and Methods

Data Collection and Processing

RNA sequencing data and corresponding clinical information for 178 pancreatic adenocarcinoma (PAAD) patients were obtained from The Cancer Genome Atlas (TCGA) database (https://portal.gdc.cancer.gov/) [22, 23]. Gene expression profiles of 171 normal pancreatic tissue samples were retrieved from the GTEx database. All expression data were log2(FPKM + 1)-transformed, and batch effects were corrected using the "limma" R package. Single-cell RNA sequencing data (GSE154778) were downloaded from the TISCH database.

Differential Expression and Survival Analyses

Differential expression of TIPARP between tumor and normal tissues was analyzed using the "limma" R package. Patients were stratified into high- and low-expression groups based on the median TIPARP expression. Kaplan–Meier survival analysis was performed to assess the prognostic relevance of TIPARP expression, and its correlations with clinical variables were evaluated [23, 24]. Univariate and multivariate Cox proportional hazards regression analyses were conducted to examine whether TIPARP served as an independent prognostic factor. A nomogram was constructed incorporating TIPARP expression and clinical variables to predict 1-, 3-, and 5-year survival probabilities. The performance of the nomogram was assessed using the concordance index (C-index) and calibration curves [25].

Immune Infiltration Analysis

To quantify immune cell infiltration, we used the "immunedeconv" R package, which integrates six state-of-the-art deconvolution algorithms. All statistical analyses were conducted using R version 4.4.3, with a significance threshold of P < 0.05. TIPARP expression was also analyzed in relation to key immunoregulatory genes and immune checkpoint molecules, including ITPRIPL1, SIGLEC15, TIGIT, CD274 (PD-L1), HAVCR2 (TIM-3), PDCD1 (PD-1), CTLA4, LAG3, and PDCD1LG2 (PD-L2) [26].

Mendelian Randomization Analysis

We performed a two-sample Mendelian Randomization (MR) analysis to assess the causal relationship between TIPARP expression and pancreatic cancer risk [27]. Summary statistics from genome-wide association studies (GWAS) for TIPARP and PAAD were obtained from the IEU OpenGWAS database (https://gwas.mrcieu.ac.uk/). Single nucleotide polymorphisms (SNPs) significantly associated with TIPARP expression (P < 1 × 10⁻⁵) and independent of each other (r² < 0.001) were selected as instrumental variables.

To ensure the robustness of the MR findings, multiple sensitivity analyses were performed, including MR-Egger intercept testing (to detect horizontal pleiotropy), Cochran’s Q test (to assess heterogeneity), and leave-one-out analysis (to evaluate the influence of individual SNPs) [28]. All MR analyses were conducted using the “TwoSampleMR” package in R (version 4.3.1), with P < 0.05 considered statistically significant [29].

Cell Culture and Transfection

Human pancreatic cancer cell lines (PANC-1 and BxPC-3) were obtained from the Cell Bank of the Chinese Academy of Sciences. PANC-1 cells were cultured in DMEM supplemented with 10% fetal bovine serum (FBS) and 1% penicillin–streptomycin. BxPC-3 cells were maintained in RPMI 1640 medium supplemented with 10% FBS, 1% L-glutamine, and 1% penicillin–streptomycin (all from Gibco®, Shanghai, China). All cells were incubated at 37 °C in a humidified atmosphere with 5% CO₂.

For transfection, cells were seeded in 6-well plates (3 × 10⁵ cells/well). When the confluence reached ~ 70%, cells were transfected with TIPARP-specific siRNAs or a TIPARP overexpression plasmid using Lipofectamine™ 3000 reagent (Thermo Fisher Scientific, USA), according to the manufacturer’s instructions. Cells were harvested 48 h post-transfection for subsequent assays.

Animal Experiments

All animal experiments were approved by the Animal Ethics Committee of the Affiliated Suzhou Hospital of Nanjing Medical University. Four-week-old male BALB/c nude mice were purchased from GemPharmatech Co., Ltd. (Jiangsu, China). Xenografts were established by subcutaneous injection of PANC-1 cells into the flanks. Subsequently, the overexpression complexes containing plasmid DNA and Lipofectamine 3000 (1:1 v/v) were locally injected into the tumor site (20 μL per mouse). Tumor volume was calculated using the formula V = (length × width²)/2. After 18 days, the mice were sacrificed, and the tumors were excised and weighed.

Functional Assays

Cell proliferation was evaluated using the Cell Counting Kit-8 (CCK-8, MCE, USA). Forty-eight hours after transfection, PANC-1 and BxPC-3 cells were seeded into 96-well plates, and cell viability was measured by detecting absorbance at 450 nm using a microplate reader (BioTek, USA).

For colony formation assays, 1,000 cells per well were seeded into 6-well plates and cultured for 7 days. Colonies were then fixed with methanol, stained with 0.1% crystal violet, and counted manually under an optical microscope.

Cell migration and invasion were assessed using Transwell chambers. Cells were seeded into the upper chamber with serum-free medium, while the lower chamber contained medium with 20% FBS. After 24 h of incubation at 37 °C, migrated cells on the lower membrane surface were fixed and stained with 1% crystal violet, then counted under a microscope.

For wound-healing assays, cells were plated in 24-well plates and grown to full confluence. A straight scratch was made using a 10 μL pipette tip. The scratch area was imaged at 0 and 12 h post-scratch to assess migration distance. All functional experiments were conducted in triplicate.

Copy Number Variation (CNV) Analysis

To assess genomic alterations, CNV data were downloaded from the UCSC Xena platform (https://xena.ucsc.edu/). The “CNVfreq” and “Rcircos” R packages were used to generate a circular plot illustrating the frequency and distribution of TIPARP-related CNV gains and losses [30].

Functional Enrichment Analysis

To explore the biological processes and signaling pathways associated with TIPARP-related genes, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses were conducted using the "clusterProfiler" and "enrichplot" R packages. Significantly enriched terms were defined by a P value < 0.05 [31].

Single-Cell Transcriptomic Analysis

Single-cell transcriptome data in.h5 format and corresponding cell-type annotation files were obtained from the TISCH database. Data preprocessing and clustering were performed using the Seurat and MAESTRO R packages. t-distributed stochastic neighbor embedding (t-SNE) was used for dimensionality reduction and visualization of cell populations.

Statistical Analysis

All statistical analyses were performed using R (version 4.0.2). For continuous variables, the Wilcoxon rank-sum test was used for non-normally distributed data; otherwise, the Student’s t-test was applied. Categorical variables were analyzed using the chi-square test or Fisher’s exact test where appropriate. Cox proportional hazards models and log-rank tests were used to evaluate hazard ratios (HRs) and 95% confidence intervals (CIs). Kaplan–Meier survival curves were plotted to visualize survival differences. A two-sided P value < 0.05 was considered statistically significant.

Results

TIPARP Expression and Prognostic Significance Across Cancers

Pan-cancer analysis revealed that TIPARP was differentially expressed in 26 out of 33 tumor types, with a significant upregulation observed in pancreatic adenocarcinoma (PAAD) (Fig. 1A). In most tumors, TIPARP expression was markedly higher in tumor tissues than in normal counterparts (P < 0.05), with particularly significant elevation in cholangiocarcinoma (CHOL), colon adenocarcinoma (COAD), esophageal carcinoma (ESCA), head and neck squamous cell carcinoma (HNSC), and PAAD (P < 0.001).

Fig. 1.

Pan-cancer expression and prognostic value of TIPARP A TIPARP mRNA expression in 33 tumor types (red) and adjacent normal tissues (blue) from the TCGA database. B–E Forest plots of the prognostic value of TIPARP expression for (B) Overall Survival (OS), C Progression-Free Interval (PFI), D Disease-Specific Survival (DSS), and E Disease-Free Interval (DFI). The size of each red circle (Hazard Ratio, HR) is proportional to the study’s weight, and the horizontal lines represent the 95% Confidence Intervals (CI). HR > 1 signifies that high TIPARP expression is a poor prognostic factor, while HR < 1 indicates a protective role. Statistical significance is achieved when the 95% CI does not overlap with the HR = 1 null effect line

Survival analysis further demonstrated the prognostic impact of TIPARP expression across multiple cancers. In overall survival (OS) analysis, high TIPARP expression was associated with poor prognosis in glioblastoma multiforme/lower-grade glioma (GBMLGG, P = 4.2e − 7), PAAD (P = 5.9e − 3), and lower-grade glioma (LGG, P = 8.8e − 3) (Fig. 1B). Similarly, progression-free survival (PFS) analysis revealed that high TIPARP expression predicted worse outcomes in GBMLGG (P = 8.8e − 11), LGG (P = 1.0e − 4), and PAAD (P = 0.01) (Fig. 1C). Disease-specific survival (DSS) and disease-free interval (DFI) analyses also supported the unfavorable prognostic role of TIPARP: higher expression was linked to shorter DSS in GBMLGG (P = 7.9e − 8), LGG (P = 5.9e − 3), and PAAD (P = 0.02) (Fig. 1D), and to earlier recurrence in PAAD (P = 0.03) (Fig. 1E).

TIPARP as an Independent Prognostic Factor and Construction of a Predictive Nomogram

To assess whether TIPARP is an independent prognostic factor in PAAD, we performed univariate and multivariate Cox regression analyses, incorporating age, sex, tumor grade, and stage. TIPARP consistently emerged as a significant independent predictor of poor prognosis (Fig. 2A, B).

Fig. 2.

Development and validation of a TIPARP-based prognostic nomogram A Univariate Cox regression analysis of prognostic factors. B Multivariate Cox regression analysis identifying independent prognostic factors. C Nomogram integrating independent prognostic factors to predict survival probability (C-index = 0.666). D Calibration curves showing the agreement between predicted and observed 1-, 3-, and 5-year survival rates. E Kaplan–Meier analysis of overall survival in patients stratified by median TIPARP expression (log-rank p = 0.008), showing that low expression correlates with worse outcomes. F Time-dependent ROC curves assessing the predictive accuracy of the nomogram at 1 year (AUC = 0.58), 3 years (AUC = 0.62), and 5 years (AUC = 0.85)

Based on these findings, we constructed a nomogram model incorporating TIPARP expression, age, and lymph node status (pN stage). Patients were stratified into high and low TIPARP expression groups based on the median value. The C-index for the model was 0.606 (95% CI: 0.544–0.668, P < 0.001), indicating moderate predictive power (Fig. 2C). The calibration curve showed good agreement between predicted and actual outcomes (Fig. 2D), and ROC analysis demonstrated high predictive accuracy at 1-, 3-, and 5-year time points (AUCs = 0.56, 0.68, and 0.885, respectively; Fig. 2F). Kaplan–Meier analysis further confirmed that patients with high TIPARP expression had significantly worse survival (P = 0.008) (Fig. 2E). Together, these results highlight the clinical utility of TIPARP in prognostic stratification.

TIPARP Expression Profile and Immunomodulatory Role in the Pancreatic Cancer Microenvironment

To dissect the cell-specific expression and immunological role of TIPARP in the PAAD tumor microenvironment, we performed integrated analyses using single-cell transcriptomics and TCGA data. t-SNE dimensionality reduction identified nine major cell types: B cells, CD4⁺ T cells, fibroblasts, epithelial cells, malignant cells, mast cells, monocytes/macrophages, myofibroblasts, and NK cells (Fig. 3A). TIPARP displayed marked spatial heterogeneity, with high expression clustered within specific cell types (Fig. 3B). Quantitative analysis revealed that fibroblasts expressed the highest levels of TIPARP, followed by myofibroblasts and NK cells; in contrast, endothelial cells, CD4⁺ T cells, and B cells had relatively low expression (Fig. 3C).

Fig. 3.

TIPARP expression in the tumor microenvironment and its correlation with immune contexture A t-SNE plot showing major cell clusters in tumor samples. B t-SNE visualization of TIPARP expression at the single-cell level. C Bar plot quantifying the average TIPARP expression in different cell subpopulations. D Heatmap comparing the abundance of 22 immune cell types between TIPARP-high and -low expression groups. E Stacked bar plot showing the cellular composition of tumor samples. F Heatmap showing a positive correlation between TIPARP expression and key immune checkpoint molecules, including CD274 (PD-L1), HAVCR2 (TIM-3), PDCD1 (PD-1), LAG3, and CTLA4 (all p < 0.05). G Correlation plots from TCGA data showing a significant positive association between TIPARP expression and the infiltration of B cells, CD4 + T cells, CD8 + T cells, neutrophils, macrophages, and dendritic cells (p < 0.05)

Using the TCGA-PAAD cohort, we stratified samples into TIPARP-high and -low groups and analyzed immune infiltration patterns. A heatmap revealed significant differences in several immune subsets between the two groups, including activated myeloid dendritic cells (P = 0.015), activated memory CD4⁺ T cells (P = 0.026), activated NK cells (P = 0.003), and resting memory CD4⁺ T cells (P = 0.035) (Fig. 3D). Differences in Tregs, CD8⁺ T cells, and macrophage subsets did not reach statistical significance. A stacked bar plot further illustrated the heterogeneity of immune infiltration across samples (Fig. 3E).

TIPARP expression was significantly positively correlated with multiple immune checkpoint molecules (Fig. 3F), including PD-L1 (CD274, P = 2.33e − 07), PD-L2 (PDCD1LG2, P = 1.12e − 03), TIM-3 (HAVCR2, P = 6.90e − 03), PD-1 (PDCD1, P = 9.15e − 03), CTLA-4 (P = 2.89e − 03), and TIGIT (P = 2.13e − 03). No significant correlations were observed for LAG3, SIGLEC15, or IGSF1.

Immune cell infiltration analysis confirmed strong correlations between TIPARP expression and infiltration levels of CD8⁺ T cells (r = 0.54, P = 8.4e − 15), neutrophils (r = 0.50), macrophages (r = 0.49), dendritic cells (r = 0.39), B cells (r = 0.27), and CD4⁺ T cells (r = 0.19) (Fig. 3G). These results suggest TIPARP is closely linked to immune landscape modulation in pancreatic cancer.

Mendelian Randomization Supports a Causal Role for TIPARP in PAAD

To explore causality, we performed two-sample Mendelian randomization (MR) using six SNPs associated with TIPARP expression (threshold: P < 1 × 10⁻⁵, R² > 0.001, Kb = 10,000, F > 10) (Fig. 4A; Table 1). Leave-one-out analysis showed no single SNP disproportionately influenced the results, and all estimates were consistently right of zero, confirming model robustness (Fig. 4B).

Fig. 4.

Mendelian randomization sensitivity analyses for pancreatic cancer risk A Forest plot comparing effect estimates from MR-Egger and Inverse-Variance Weighted (IVW) analyses. B Leave-one-out sensitivity analysis using the IVW method, where the red dashed line shows the overall causal estimate. C Funnel plot assessing pleiotropic bias, showing results from MR-Egger (blue), IVW (green), Simple Mode (gray), and Weighted Median (red dashed line) methods. The axes represent SNP-exposure effects (βX) versus the inverse of their standard errors (1/SE). D Scatter plot illustrating the relationship between SNP effects on the exposure (βX) and SNP effects on pancreatic cancer risk (βY), with the MR-Egger regression line shown in blue

Table 1.

Association between TIPARP and pancreatic cancer GWAS

SNP	Exposure			Outcome
	Beta	SE	P	Beta	SE	P
rs114679164	−0.359205	0.0482621	9.86e-14	−0.1632	0.1535	0.2878
rs13164856	0.0585965	0.0132019	9.06e-06	0.0375	0.0381	0.3258
rs344059	0.368493	0.0294453	6.21e-36	0.1997	0.1033	0.0530701
rs55793580	−0.167173	0.015788	3.37e-26	−0.0797	0.0448	0.0754606
rs56205728	0.058523	0.0129639	6.36e-06	−0.0051	0.0454	0.9099
rs597808	0.0904245	0.0118926	2.89e-14	−0.0097	0.0415	0.8145

MR-Egger intercept test (P = 0.484) and Cochran’s Q test (Q = 0.826 for MR-Egger; Q = 0.835 for IVW) showed no evidence of pleiotropy (Table 2). A funnel plot confirmed symmetrical distribution, indicating minimal bias (Fig. 4C). The inverse-variance weighted (IVW) method revealed a significant causal relationship between TIPARP and PAAD risk (OR = 1.511, 95% CI: 1.112–2.053, P = 0.008) (Table 3; Fig. 4D), supporting TIPARP as a potential oncogenic driver.

Table 2.

The Egger regression test evaluates potential multivariate interaction effects

Pleiotropy			Heterogeneity
MR_Egger			MR Egger			IVW
Intercept	SE	P	Q	Q_df	Q_p	Q	Q_df	Q_p
-0.02712	0.03519	0.484	1.506	4	0.826	2.100	5	0.835

Table 3.

Six methodologies were employed to evaluate the causal association between TIPARP and the susceptibility to PAAD

Method	SNP(n)	OR	95% CI	P
MR Egger	6	1.789	1.06–3.03	0.097
WM	6	1.623	1.11–2.3	0.013
IVW	6	1.511	1.1–2.05	0.008
Simple mode	6	1.677	1.02–2.75	0.095
Weighted mode	6	1.654	1.09–2.51	0.065

TIPARP Knockdown Inhibits Proliferation and Migration of Pancreatic Cancer Cells

Using two independent siRNAs (si-1 and si-2), we silenced TIPARP in PANC-1 and BxPC-3 cells. CCK-8 assays revealed significantly reduced proliferation in TIPARP-silenced cells compared to controls at 24 h, 48 h, and 72 h (Fig. 5A). Colony formation assays showed both fewer and smaller colonies upon TIPARP knockdown, particularly in BxPC-3 cells (Fig. 5B). Transwell migration and wound-healing assays further confirmed that TIPARP silencing impaired the migratory and invasive capacity of pancreatic cancer cells (Fig. 5C–D), highlighting its role in tumor aggressiveness.

Fig. 5.

TIPARP knockdown impairs the malignant phenotypes of pancreatic cancer cells Silencing TIPARP expression suppressed the proliferation (A), colony formation (B), invasion (C), and migration (D) of PANC-1 and BxPC-3 cells. Results are shown as mean ± SD from three independent replicates. *P < 0.05, **P < 0.01, ***P < 0.001

TIPARP Overexpression Enhances Proliferation and Migration

To verify TIPARP’s oncogenic potential, we constructed an overexpression vector (OE). CCK-8 and colony formation assays demonstrated significantly enhanced proliferation in TIPARP-overexpressing PANC-1 and BxPC-3 cells (Fig. 6A–B). Transwell and wound-healing assays confirmed increased migratory and invasive abilities in the OE group compared to controls (Fig. 6C–D), functionally validating. In xenograft models, TIPARP-overexpressing cells formed significantly larger and heavier tumors than controls (Fig. 6E), confirming the tumor-promoting role of TIPARP in vivo.

Fig. 6.

TIPARP overexpression enhances the malignant phenotypes of pancreatic cancer cells. TIPARP overexpression promoted the proliferation (A), colony formation (B), invasion (C), and migration (D) of PANC-1 and BxPC-3 cells. Results are shown as mean ± SD from three independent replicates. E In vivo xenograft assays showing that TIPARP overexpression significantly increased tumor growth rate and final tumor weight *P < 0.05, **P < 0.01, ***P < 0.001

Functional Enrichment of TIPARP-Related Genes

To explore downstream pathways, transcriptomic analysis identified 1,499 differentially expressed genes (DEGs) between TIPARP-high and -low groups (|log₂FC|> 1, FDR < 0.05), including 847 upregulated and 652 downregulated genes (Fig. 7A). Circular heatmap visualization highlighted the genomic features of TIPARP in PAAD (Fig. 7B).

Fig. 7.

Gene expression changes and functional enrichment analysis. A Volcano plot showing differentially expressed genes, with x-axis representing log2(1.5-fold change) and y-axis representing -log10(P-value). Blue and red dots indicate significantly upregulated and downregulated genes, respectively, with dashed lines representing statistical significance thresholds (P < 0.05). B Circular plot analysis displaying gene expression profiles of TIPARP in pancreatic adenocarcinoma, with colors ranging from blue (Exp = −2) to red (Exp = 2) indicating expression level changes. The circular structure shows associations or co-expression patterns between genes, with inner and outer circles representing gene names. C GO functional enrichment bar chart showing enriched GO terms and their gene counts, with red indicating biological processes, blue indicating molecular functions, and orange indicating cellular components. X-axis represents gene count and y-axis represents GO term names. D Scatter plot showing KEGG pathway enrichment results, with x-axis representing enrichment factor (-log10(P-value)) and y-axis representing pathway names. Dot size indicates gene count, colors from blue to red represent enrichment significance, and circles represent specific pathways

Gene Ontology (GO) enrichment analysis revealed that TIPARP-related DEGs were involved in extracellular matrix organization, vasculature development, immune response, and cell adhesion (Fig. 7C). KEGG analysis identified enrichment in key signaling pathways including PI3K-Akt, ECM-receptor interaction, focal adhesion, cytokine–cytokine receptor interaction, chemokine signaling, TNF signaling, and IL-17 signaling (Fig. 7D), collectively pointing toward TIPARP’s involvement in tumor progression, immune modulation, and stromal remodeling.

Discussion

Our pan-cancer analysis reveals that TIPARP is significantly upregulated in pancreatic adenocarcinoma (PAAD) and strongly correlates with multiple prognostic indicators, including overall survival (P = 5.9 × 10⁻³), progression-free survival (P = 0.01), disease-specific survival (P = 0.02), and disease-free interval (P = 0.03). These findings align closely with those reported by Gozgit et al. [8], who observed that elevated PARP7 expression enhances tumor cell aggressiveness and resistance to apoptosis across various tumor cell lines.

Multivariate Cox regression analysis further confirmed TIPARP as an independent prognostic risk factor, irrespective of age, sex, or tumor grade. Based on this, we developed a nomogram incorporating TIPARP expression, age, and nodal (pN) stage, which demonstrated robust predictive performance (C-index = 0.666; AUC = 0.885 for 5-year survival), outperforming traditional clinical models. This superiority may stem from TIPARP’s capacity to more directly reflect the biological behavior of tumors as a functional molecular marker. Supporting this, Cohen et al. [7] highlighted that ADP-ribosylation, a key post-translational modification mediated by PARP family proteins, rapidly responds to cellular stress and modulates various biological pathways—underpinning TIPARP’s prognostic relevance.

Compared to previous research, our study provides novel contributions by validating TIPARP’s prognostic value in a large patient cohort and by developing a clinically applicable predictive model. These results not only enhance current risk stratification frameworks for PAAD but also offer a foundation for personalized therapeutic decision-making.

To better understand TIPARP’s biological role and its prognostic significance, we explored its distribution in the tumor microenvironment using single-cell transcriptomic data. TIPARP was predominantly expressed in fibroblasts, followed by myofibroblasts and NK cells. Its enrichment in cancer-associated fibroblasts (CAFs) is particularly notable, as these cells are central to the dense stromal architecture characteristic of PAAD [32]. High TIPARP expression in CAFs may contribute to maintaining pathological fibrosis and regulating stromal responses. As noted by Furukawa et al. [15], this desmoplastic stroma not only serves as a structural hallmark of pancreatic cancer but also impedes drug delivery and immune cell infiltration. Our findings suggest that TIPARP may act as a key molecular mediator in this process.

Analysis of the immune microenvironment further revealed a strong positive correlation between TIPARP and multiple immune checkpoint molecules, most notably CD274 (PD-L1) (P = 2.33 × 10⁻⁷). Gao et al. [33] have shown that PARP7 suppresses type I interferon signaling, thereby dampening immune activation. Given PD-L1’s pivotal role in tumor immune evasion, the synergy between TIPARP and PD-L1 may underlie the formation of an immunosuppressive microenvironment in PAAD. Popova et al. [34] further demonstrated that PARP7 regulates transcription factors such as FRA1 and NF-κB/RelA through ADP-ribosylation, providing mechanistic insight into these immunomodulatory effects.

Interestingly, we also observed a significant positive correlation between TIPARP expression and CD8⁺ T cell infiltration (r = 0.54, P = 8.4 × 10⁻¹⁵). While seemingly paradoxical, this may reflect the complexity of the tumor-immune interaction. Although CD8⁺ T cells are recruited, their effector function could be compromised by co-existing immunosuppressive signals (e.g., PD-L1 expression, SOCS3-mediated JAK/STAT inhibition), leading to T cell exhaustion or a compensatory immune activation state [35]. This aligns with the immune evasion model described by Laheru et al. [20], and our data establish TIPARP as a potential central regulator within this immunosuppressive network.

Our Mendelian randomization (MR) analysis provides strong genetic evidence supporting a causal role of TIPARP in pancreatic tumorigenesis—an insight that observational studies alone cannot provide. Using six independent instrumental SNPs, we found that genetically elevated TIPARP expression was associated with a 51% increased risk of PAAD (OR = 1.511, P = 0.008). This causal relationship may be attributable to TIPARP’s role in modulating cellular stress responses—such as DNA damage repair, oxidative stress adaptation, and metabolic regulation—as highlighted by Cohen et al. [7]. By promoting stress resilience, TIPARP may facilitate the malignant transformation of pancreatic epithelial cells. This approach exemplifies the principle advocated by Fong et al. [12], wherein causal validation is essential for identifying true oncogenic drivers. Moreover, the consistency across multiple sensitivity analyses strengthens the reliability of our MR findings.

To experimentally validate TIPARP’s functional role, we conducted in vitro assays in pancreatic cancer cell lines. TIPARP knockdown markedly inhibited cell proliferation, colony formation, and migration, whereas overexpression yielded opposite effects. Specifically, PARP7 overexpression robustly enhanced the proliferation, clonogenicity, migration, and invasion of pancreatic cancer cells. Crucially, these pro-oncogenic phenotypes were all significantly attenuated by the selective PARP7 inhibitor RBN-2397 (500 nM). These phenotypic changes are consistent with our pathway enrichment analysis, which showed TIPARP-associated genes were enriched in the PI3K-Akt, ECM-receptor interaction, and focal adhesion pathways. TIPARP may co-regulate these pathways by modulating transcription factors like FRA1, thereby facilitating tumor progression [10].

The PI3K-Akt pathway promotes cell survival and proliferation, while ECM-receptor and focal adhesion signaling are central to cell adhesion, motility, and invasion [36, 37]. TIPARP’s ability to coordinate the activation of these pro-oncogenic pathways highlights its distinct role from other PARP family members (e.g., PARP1/2), which are primarily involved in DNA repair. Notably, TIPARP regulates transcriptional activity through mono-ADP-ribosylation, offering a unique mechanism and a promising target for cancer therapy [38, 39].

Our findings also identified co-expression between TIPARP and genes involved in ECM remodeling and angiogenesis, offering new perspectives on its role in pancreatic cancer invasion and metastasis. Given TIPARP’s multifaceted regulatory functions, it presents an attractive candidate for therapeutic intervention.

Conclusion

In conclusion, this study identifies TIPARP as a novel oncogenic driver and key regulator of the tumor immune microenvironment in pancreatic adenocarcinoma. Through comprehensive pan-cancer, transcriptomic, single-cell, and Mendelian randomization analyses, as well as in vitro functional assays, we demonstrate that TIPARP is not only an independent prognostic biomarker but also a potential therapeutic target. Mechanistically, TIPARP promotes tumor progression by activating oncogenic signaling pathways (e.g., PI3K-Akt) and fostering immune evasion (e.g., via PD-L1 upregulation and CD8⁺ T cell exhaustion).

Our findings provide critical insight into TIPARP’s role in PAAD and establish a solid foundation for developing TIPARP-targeted therapies—either as monotherapy or in combination with immunotherapy (e.g., RBN-2397, (S)-XY-05). Further research is warranted to validate and translate these discoveries into clinical applications that improve outcomes for patients with pancreatic cancer.

Author Contributions

HJ: Writing–review & editing, Writing–original draft, Visualization, Validation, Software, Resources, Project administration, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. WN: Validation, Methodology, Data curation. YP: Supervision, Funding acquisition. ZJ: Review, Editing. ZQ: Supervision, Funding acquisition. JW: Resources. XJ: Supervision, Funding acquisition. LH: Project administration, Conceptualization.

Funding

This work was supported by Suzhou Clinical Medical Center Construction Project (Szlcyxzxj202107), Suzhou City "Science and Education Promoting Health" Youth Science and Technology Project (KJXW2023040), Suzhou City Applied Basic Research (Medical and Health) Technology Innovation Project(SYW2024028) and Suzhou City Applied Basic Research (Medical and Health) Technology Innovation Project(SKY2022192).

Data Availability

The datasets analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Conflict of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.