p53 and PD-1/PD-L1 expression contribute to disease progression and are correlated to immune infiltrates in urothelial carcinoma

Abstract

The relevance and clinical significance of p53 and PD-1/PD-L1 in urothelial carcinoma (UC) are still unknown. This study was to explore the expression, clinical significance, and correlation of p53, PD-1/PD-L1, as well as their associations with immune cells, immune checkpoints and immunotherapy in UC. The expression of p53 and PD-1/PD-L1 were analyzed by the tumor immune estimation resource (TIMER), SangerBox, Gene Expression Profiling Interactive Analysis (GEPIA) databases and immunohistochemistry. The University of ALabama at Birmingham CANcer data analysis Portal (UALCAN) and Kaplan-Meier plotter databases were used to examine the clinical and prognostic value of p53 and PD-1/PD-L1 in bladder cancer (BLCA). Next, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses were performed to analyze the potential mechanisms between p53, PD-1/PD-L1, and their co-expression genes and proteins identified through the GeneMANIA, STRING databases, and competitive endogenous RNA (ceRNA) network. The TIMER, GEPIA2, and An integrated repository portal for tumor-immune system interactions (TISIDB) databases were used to analyze the correlation of p53 and PD-1/PD-L1 expression with immune cell infiltration and immune cell gene markers in BLCA. Finally, the association between p53, PD-1/PD-L1 expression and immunotherapy checkpoint inhibitor (ICB), tumor mutation burden (TMB), tumour immune dysfunction and exclusion (TIDE) scores, and immunotherapy in TCGA-BLCA data was analyzed using the “Limma” package. Overall, p53 and PD-L1 expression were found to be significantly different between UC tissues and adjacent normal tissues, whereas no significant difference in PD-1 expression was observed. Pan-cancer survival analysis showed that p53, PD-1/PD-L1 were significantly associated with the prognosis of a variety of pan-cancers, including overall survival (OS) and relapse-free survival (RFS). However, further analysis also confirmed that only low PD-1 expression was associated with poorer OS and RFS in BLCA. In addition, p53 and PD-1/PD-L1 expression are closely related to adverse clinicopathological features. The correlation analysis between p53 and PD-1/PD-L1 showed a significant negative correlation between p53 and PD-1, while PD-1 was significantly positively correlated with PD-L1. Notably, p53 and PD-1/PD-L1 were found to be involved in the regulation of immune responses in GeneMANIA, STRING, ceRNA network, and functional enrichment analysis. Further analysis indicated that p53 and PD-1/PD-L1 were associated with specific immune cells and immune cell gene markers, which may partially affect UC prognosis due to the level of immune cell infiltration. Meanwhile, the correlation analysis of p53, PD-1/PD-L1 with ICB, TMB, TIDE scores and immunotherapy revealed that p53 had a better immunotherapeutic effect in PD-1 negative BLCA patients; Whereas, high PD-1/PD-L1 expression had a better immunotherapeutic effect regardless of CTLA4 and/or PD-1 positivity. As an immune gene or protein associated with PD-1/PD-L1, p53 is significantly negatively correlated with PD-1. High expression of p53 may inhibit PD-1 expression, further inhibiting the PD-1/PD-L1 axis to reduce immunosuppressive status. In addition, p53 may also block the formation of PD-1/PD-L1 resistance by inhibiting the polarization of TAMs and M2 macrophages. This work demonstrates the important role of p53 in PD-1/PD-L1 axis-based immunotherapy for UC patients, and p53 is expected to become a key target for breaking through the current status of immunotherapy.

Introduction

Urothelial carcinoma (UC) is not only a common malignant tumor of the urinary system, but also the most common histological type, which can occur anywhere in the renal pelvis, ureter, bladder, and urethra. Its morbidity and mortality are increasing year by year, and it has become the fourth malignant tumor in men^1,2. According to the location of the tumor, UC can be divided into upper and lower urinary tract urothelial carcinomas. Lower urinary tract urothelial carcinoma (LTUC) is mainly bladder cancer (BLCA), which accounts for 90–95% of all UC, with a high incidence and postoperative recurrence rate^3,4. Upper urinary tract urothelial carcinoma (UTUC), which includes the renal pelvic carcinoma and the ureteral carcinoma, accounts for only 5–10% of all UC, with an incidence of about 2/1000,000 in Europe and the United States⁵. However, the mutation rate is high, and it has the characteristics of a hidden disease, strong invasiveness, a high recurrence rate, and poor prognosis^6,7. Up to now, although significant progress has been made in the clinical treatment of early UC, locally advanced and metastatic urothelial carcinoma is still very difficult to treat, are currently incurable, and still have a poor prognosis^8,9. Therefore, further exploration of its carcinogenic mechanism and potential drug targets is crucial for UC patients.

The TP53 gene is the most common gene in human cancers, and about 50% of human cancers are related to its mutations¹⁰. It encodes the tumor suppressor protein p53, which acts as a key transcription factor to inhibit tumor growth processes, including cell cycle, DNA replication, and uncontrolled cell division¹¹. However, p53 dysfunction can promote inflammation and support tumor immune evasion, thereby becoming an immune driving factor for tumor occurrence¹². In addition, human PD-1 (CD279, PDCD1), encoded by the PDCD1 gene, is an immunoglobulin superfamily member. Its ligand, PD-L1 (B7-H1, CD274), belongs to the B7 protein family. In healthy hosts, PD-1/PD-L1 signaling regulates T cells to respond and protect normal tissue from immune-mediated damage, induce immune tolerance, and avoid autoimmune diseases¹³. In tumor immunity, PD-L1 binds to PD-1 expressed on cytotoxic T lymphocytes (CTL) and evades immune surveillance by inducing CTL apoptosis¹⁴. Recent studies have shown that p53 and PD-1/PD-L1 are expressed in both LTUC and UTUC patients and are potential prognostic markers^{15, 16, 17–18}. However, the potential mechanisms of p53 and PD-1/PD-L1 in the progression, prognosis, and immune infiltration of UC are not yet clear, and their potential correlations in UC is not yet clear.

In this article, we focused on BLCA as the research topic for UC and comprehensively analyzed the expression of p53 and PD-1/PD-L1 in pan-cancer by using several public databases. Besides, we explored the correlation of p53 and PD-1/PD-L1 with clinicopathological features, prognosis, immune cell infiltration, immune cell markers, and immunotherapy in BLCA patients based on various databases. Then, the protein expression levels of p53 and PD-1/PD-L1 in UC patients and adjacent normal tissues were preliminarily verified by immunohistochemical analysis, and we also verified their correlation and their relationship with clinicopathological parameters. Meanwhile, this study preliminarily explored the potential value of p53 as a therapeutic target for PD-1/PD-L1 axis, in order to provide a theoretical basis for targeted treatment and monitoring of UC.

Materials and methods

Patients’ selection and tissue collection

From January 2013 to December 2020, 65 cases of urothelial carcinoma tissues (29 UTUC and 36 BLCA) and 31 cases of adjacent normal tissues were collected for immunohistochemistry at the Affiliated Hospital of Zunyi Medical University, Zunyi, Guizhou Province, China. All patients were histologically diagnosed with urothelial carcinoma and did not receive chemotherapy or radiotherapy before surgery. Clinical characteristics such as age, gender, tumor diameter, tumor multiplicity, tumor grade, tumor stage, and lymph node metastasis were obtained from our hospital clinical database for all patients. Transcriptome data and RNA-Seq data of 431 BLCA cases, including 19 normal samples and 412 tumor samples, were obtained from The Cancer Genome Atlas Program (TCGA) database (https://portal.gdc.cancer.gov/). All experiments were performed according to the principles of the Declaration of Helsinki, and written informed consent was obtained from all patients. The study was approved by the Institutional Ethics Committee of the Affiliated Hospital of Zunyi Medical University (ethics no.KLLY-2022-225).

Immunohistochemistry and analysis

Formalin-fixed paraffin-embedded UC tissue was cut into 4-µm-thick serial sections, which were baked in an oven at 60 °C for 2 h. After deparaffinization in microwave-heated antigen uncovering solution, rehydration, and microwave heating to repair the antigen, the sections were blocked with a shared goat serum for 30 min. Next, the sections were incubated with anti-p53 monoclonal antibody (Zenbio, 345567, Abcam, 1:100 dilution), anti-PD-1 monoclonal antibody (Zenbio, R50010, Abcam, 1:200 dilution), and anti-PD-L1 monoclonal antibody (Servicebio, GB11339A, Abcam, 1:1000 dilution) were incubated at 4 °C overnight. Afterwards, the HRP-coupled secondary antibody (Boster) was incubated for 50 min at room temperature. Goat serum was used as a negative control. Then, tissue sections were restained with hematoxylin, dehydrated, and mounted on coverslips.

Finally, the staining results were evaluated by two pathologists who were not aware of the clinical parameters. Immunohistochemical staining was assessed for p53. The following scoring method was used: 25%, 0 points; 26–50%, 1 point; 51–75%, 2 points; and > 75%, 3 points. The staining intensity was categorized as follows: Uncolored, 0 points; pale yellow, 1 point; yellow or brown, 2 points; and brown, 3 points. According to the final points, 0–1 was (−), 2–3 points were weakly positive (+), 4–5 points were positive (++), and > 5 were strongly positive (+++). Specimens that scored (+) to (+++) were considered high expression, while samples scored from (-) are considered low expression¹⁹. Immunohistochemical staining for PD-L1, and PD-1 was then performed. Membranous positivity in tumor cells and any cytoplasmic/membranous staining in mononuclear cells were considered as high expression for PD-L1. PD-L1 immunohistochemistry was scored as the percentage of positive cells. The cases were analyzed using a cut-off of both 1% and 5% for both lymphocytes and tumor cells separately. PD-1 was evaluated in mononuclear cells, and any cytoplasmic/membranous positivity was taken as high expression. The staining percentage was calculated on the slides examined for the immunohistochemistry by calculating the percentage of lymphocytes or tumor cells showing staining^20,21.

SangerBox database analysis

SangerBox (http://SangerBox.com/Tool) database is a useful and free online platform for TCGA database analysis²². We entered “TP53”, “PDCD1” and “CD274” respectively in this web server to investigate the expression differences between tumors and adjacent normal tissues from datasets in Genotype-Tissue Expression (GTEx) and TCGA databases. TP53, PDCD1 and CD274 gene expression data were extracted for each sample and transformed using log2(X + 0.001), and tumor species with fewer than 0 samples were excluded from the analysis.

UALCAN database analysis

UALCAN (http://ualcan.path.uab.edu/index.html)database is an open online analysis tool that provides comprehensive cancer transcriptome and clinical patient data²³. In our study, we not only evaluated the expression of p53 and PD-1/PD-L1 using the “Expression Analysis” module, but also investigated the correlation between the expression of p53 and PD-1/PD-L1 and clinicopathological features, including age, gender, cancer stage, tumor histology, molecular subtypes, and lymph node metastasis. In addition, we also evaluated the relationship between p53, PD-1/PD-L1, and BLCA prognosis through the “survival analysis” module. The UALCAN platform derives median expression values from graphical plots and performs a Student’s t-test to evaluate the statistical significance of observed differential expression between normal and neoplastic samples.

GAPIA2 database analysis

GAPIA2 (http://gepia2.cancer-pku.cn/) database is a convenient and developed bioinformatics online analysis platform for analyzing RNA sequencing expression data from TCGA and GTEx databases using standard processing pipelines²⁴. In this article, the GEPIA database is used as a supplement to address the shortcomings of the TIMER database. In addition, the database also explored the differential expression of p53, PD-1/PD-L1 between BLCA tissues and adjacent normal tissues, as well as their correlation. For box plot visualization, expression values were normalized using log2(TPM + 1) transformation. Differential analysis in GEPIA2 was performed using the LIMMA method, with a fold change threshold of |log2FC| ≥1 and a significance cutoff of q-value < 0.01.

GeneMANIA database analysis

GeneMANIA website (http://genemania.org) used to predict functionally similar genes of hub genes and construct a network between them²⁵. This study investigated the 20 common functionally similar co-expression genes of p53 and PD-1/PD-L1, and constructed a gene-gene interaction network of p53 and PD-1/PD-L1.

STRING database analysis

The STRING database (https://string-db.org) is a useful online analysis tool for evaluating protein interactions in multiple ways²⁶. Based on the co-expression genes in the GeneMANIA database, we set the species as “human” and used a composite score > 0.4 as the threshold for inclusion in the network, and constructed a protein-protein interaction network for p53 and PD-1/PD-L1 in this study.

Construction of the CeRNA network for key genes

Potential regulatory microRNAs (miRNAs) targeting the key genes of interest were predicted using four well-established miRNA prediction databases: miRDB, miRanda, miRWalk, and TargetScan. To ensure high-confidence predictions, only miRNAs consistently identified across all four databases were retained as candidate regulatory miRNAs. Following miRNA identification, we proceeded to predict miRNA-interacting long non-coding RNAs (lncRNAs) using the SpongeScan database, which specializes in identifying potential miRNA sponges. These computationally predicted interactions were then integrated to construct a competitive endogenous RNA (ceRNA) network, representing the complex mRNA-miRNA-lncRNA regulatory relationships. The network visualization and analysis were performed using Cytoscape software (version 3.10.1), enabling comprehensive topological examination of the ceRNA interactions.

GO analysis and KEGG analysis

DAVID(https://david.ncifcrf.gov/)online analysis tool conducted Gene Ontology (GO)²⁷ and Kyoto Encyclopedia of Genes and Genomes (KEGG)^28,29 analysis on the genes interacting with p53 and PD-1/PD-L1. GO analysis includes gene annotation results of biological processes (BPs), cellular components (CCs), and molecular functions (MFs). The results of GO analysis and KEGG analysis are further visualized using online visualization tools (http://www.bioinformatics.com.cn).

TIMER database analysis

Tumor Immune Estimation Resource (TIMER) (https://cistrome.shinyapps.io/timer/) database is a synthetic network that was used to analyze the immune infiltration of different cancers in the TCGA database³⁰. In this work, we used the “Diff Exp module” to confirm the expression of p53 and PD-1/PD-L1 in different tumors, respectively. Then, the “Gene module” was used to estimate the relationship between p53, PD-1/PD-L1 and immune cell infiltration levels, including CD4 T cells, CD8 T cells, B cells, macrophages, dendritic cells (DC) and neutrophils. In addition, the “SCNA module” was used to compare the tumor immune cell infiltration levels of different somatic copy number changes in p53 and PD-1/PD-L1 in BLCA. Wilcoxon rank sum test was used to evaluate the differences in infiltration levels between each category of somatic copy number changes (SCNA) and normal levels, including deep-deletion, arm-level deletion, diploid/normal, arm-level gain, and high amplification. Finally, using the “Correlation module”, we investigated the relationship between p53, PD-1/PD-L1 and genetic markers of immune infiltration cells in BLCA, including tumor-associated macrophages (TAM), M1 macrophages, M2 macrophages, DCs, neutrophils, T-helper 1 (Th1) cells, T-helper 2(Th2) cells, regulatory T cells (Tregs), natural killer (NK) cells, B cells, CD8 + T cells, CD4 + T cells and T cell exhaustion. P < 0.05 was considered statistically significant, *P < 0.05, **P < 0.01, ***P < 0.001.

TISIDB database analysis

TISIDB (http://cis.hku.hk/TISIDB/index.php/) is an online analysis website for the interaction between genes and tumor immunity³¹. The expression of p53 and PD-1/PD-L1 genes in different immune subtypes was analyzed, including C1 (wound healing), C2 (IFN-gamma Dominant), C3 (inflammatory), C4 (lymphocyte-depleted), and C6 (TGF-β Dominant) subtypes. Associations were assessed using Pearson’s correlation coefficient, with a statistical significance threshold of P < 0.05.

Kaplan-Meier plotter database

Kaplan−Meier plotter (http://kmplot.com) database is an online database containing gene expression data and survival information³². In this article, it is used to analyze the relationship between the mRNA expression levels of p53 and PD-1/PD-L1 and survival of BLCA, including overall survival (OS) and relapse-free survival (RFS). Patients were classified into low- and high-expression groups based on median expression values, determined through the ‘Auto select best cut-off’ algorithm. At the same time, we also calculated the hazard ratio (HR) with a 95% confidence interval and logarithmic rank p-value. P < 0.05 was considered statistically significant, statistical significance was assessed using uncorrected p-values (no multiple testing adjustment applied).

Tumor immunological analysis of TCGA-BLCA data

According to the median expression levels of p53 and PD-1/PD-L1, the data were divided into high expression group and low expression group. The correlations among p53, PD-1/PD-L1 expression, immune checkpoint inhibitors (ICBs) response, tumor mutation burden (TMB), tumor immune dysfunction and exclusion and (TIDE) scores, and immunotherapy efficacy were evaluated in BLCA patients using the ‘Limma’ package. 47 commonly studied immune checkpoint markers were selected based on previous studies. Pearson correlation coefficients were calculated to assess the associations of p53, PD-1/PD-L1 and these markers, followed by analysis of their relationships with TMB. Moreover, Immunophenoscores (IPS) for BLCA were obtained from The Cancer Immunome Atlas (TCIA) (https://tcia.at/). We evaluated the association of p53 and PD-1/PD-L1 expression with IPS to predict immunotherapy sensitivity. Additionally, the TIDE algorithm (http://tide.dfci.harvard.edu/) was employed to compute TIDE scores, which estimate patient responsiveness to immune checkpoint blockade (ICB) therapies targeting PD-L1, PD-1, and CTLA4. The limma and ggpubr packages were used to visualize the differences in TIDE scores between high and low expression groups of P53, PD-1/PD-L1. The statistical significance threshold is P < 0.05.

Statistical analysis

Statistical data were analyzed using R 4.3.2 and SPSS 29.0. Differential expression of different subgroups was analyzed by Student’s t-tests or Wilcoxon rank-sum tests or log rank tests. χ2-test and Fisher’s exact test were used to analyze the correlation between p53 and PD-1/PD-L1 expression and clinicopathological parameters. Spearman’s correlation and statistical significance were used to evaluate the correlation of p53 and PD-1/PD-L1 expression, as well as their correlation with tumor immune cells, respectively. The strength of the correlation is shown as R-values: 0-0.30, “weak”; 0.31–0.60, “moderate”; 0.61–0.80, “strong”; 0.81-1.00, “very strong “. P < 0.05 was considered statistically significant, *P < 0.05, **P < 0.01, ***P < 0.001.

Results

Expression patterns of p53 and PD-1/PD-L1 in human tumors

To establish baseline expression profiles and identify potential roles in oncogenesis and immune regulation, we first characterized p53 and PD-1/PD-L1 expression across human cancers. Using the TIMER database, we conducted a comprehensive analysis of p53 and PD-1/PD-L1 mRNA expression levels across various cancer types. Due to the limited availability of normal tissue samples in TIMER, we supplemented this analysis with data from the GEPIA database. Our results demonstrated that p53 mRNA expression was significantly elevated in multiple tumor types compared to corresponding normal tissues, including BLCA, cervical and endocervical cancer (CESC), cholangiocarcinoma (CHOL), colon adenocarcinoma (COAD), esophageal carcinoma (ESCA), glioblastoma multiforme (GBM), head and neck cancer-HPVpos (HNSC-HPVpos), kidney chromophobe (KICH), kidney renal clear cell carcinoma (KIRC), kidney renal papillary cell carcinoma (KIRP), liver hepatocellular carcinoma (LIHC), lung adenocarcinoma (LUAD), lung squamous cell carcinoma (LUSC), prostate adenocarcinoma (PRAD), rectum adenocarcinoma (READ), stomach adenocarcinoma (STAD), testicular germ cell tumors (TGCT), thyroid carcinoma (THCA), diffuse large B-cell lymphoma (DLBC), acute myeloid leukemia (LAML) and uterine corpus endometrial carcinoma (UCEC) (Fig. 1A, D). Similarly, PD-1 expression was significantly upregulated in tumor tissues of breast invasive carcinoma (BRCA), CHOL, ESCA, GBM, HNSC, HNSC-HPVpos, KICH, KIRC, KIRP, LIHC, LUAD, sarcoma (SARC), STAD, TGCT, THCA, UCEC, DLBC and LAML compared to normal controls (Fig. 1B, E). PD-L1 overexpression was observed in BRCA, CESC, CHOL, ESCA, HNSC, LIHC, LUAD, LUSC, PRAD, skin cutaneous melanoma (SKCM), STAD, UCEC and DLBC (Fig. 1C, F). These findings were further validated using the SangerBox database, which showed consistent pan-cancer expression patterns for all three genes (Fig. 1G-I).

Fig. 1 [Images not available. See PDF.]

The expression levels of p53 and PD-1/PD-L1 mRNA in different types of human cancer. (A-C) p53 and PD-1/PD-L1 expression in different cancer datasets from the TIMER database. (D-F) Supplement to the GEPIA database on the expression of p53 and PD-1/PD-L1 in partial cancer datasets. (G-I) p53 and PD-1/PD-L1 expression in pan-cancer in the SangerBox database. *P < 0.05, **P < 0.01, ***P < 0.001.

In UC specifically, both UALCAN and GEPIA2 database analyses confirmed that p53 and PD-L1 mRNA expression were significantly increased in tumor tissues compared to adjacent normal tissues, while PD-1 expression showed no significant difference (Fig. 2A-F). Immunohistochemical staining of 65 UC tissues and 31 corresponding normal tissues corroborated these findings at the protein level (Fig. 2G) (Table 1). Detailed subgroup analysis revealed that p53 and PD-L1 protein expression were significantly elevated in both BLCA and UC tissues compared to normal controls, while PD-1 expression remained unchanged (Table 2) (Supplementary table S1). Interestingly, no significant differences were observed in UTUC (Supplementary table S2). These results establish tumor-specific expression patterns, with p53 and PD-L1 showing consistent overexpression in UC, suggesting their potential involvement in UC pathogenesis and immune regulation.

Fig. 2 [Images not available. See PDF.]

The expression of p53 and PD-1/PD-L1 in UC cancer tissues and adjacent normal tissues. (A-C) p53 and PD-1/PD-L1 mRNA expression in BLCA tissues and adjacent normal tissues was examined by using UALCAN database. (D-F) p53 and PD-1/PD-L1 mRNA expression was examined by using GEPIA2 database in BLCA. (G) p53 and PD-1/PD-L1 protein expression was examined in UC cancer tissues and unpaired adjacent normal tissues by IHC staining (400×, scale = 50 μm). *P < 0.05, **P < 0.01, ***P < 0.001.

Table 1. Expression of p53 and PD-1/PD-L1 in UTUC, BLCA and unpaired adjacent normal tissues.

Groups	p53		Total	PD-1		Total	PD-L1		Total
	Low	High		Low	High		Low	High
Adjacent normal tissue	17 (54.8%)	14 (45.2%)	31	23 (74.2%)	8 (25.8%)	31	25 (80.6%)	6 (19.4%)	31
UTUC	13 (44.8%)	16 (55.2%)	29	21 (72.4%)	8 (27.6%)	29	18 (62.1%)	11 (37.9%)	29
BLCA	8 (22.2%)	28 (77.8%)	36	24 (66.7%)	12 (33.3%)	36	19 (52.8%)	17 (47.2%)	36
Total	38	58	96	68	28	96	62	34	96

Table 2. Expression of p53 and PD-1/PD-L1 in BLCA tissues and unpaired adjacent normal tissues.

Groups	Case No.	p53		P	PD-1		P	PD-L1		P
		Low	High		Low	High		Low	High
Adjacent normal tissue	31	17 (54.8%)	14 (45.2%)	0.006	23 (74.2%)	8 (25.8%)	0.502	25 (80.6%)	6 (19.4%)	0.017
BLCA	36	8 (22.2%)	28 (77.8%)		24 (66.7%)	12 (33.3%)		19 (52.8%)	17 (47.2%)

Prognostic analysis and clinicopathological correlations of p53 and PD-1/PD-L1 in human tumors

To evaluate clinical relevance, we examined associations between marker expression, patient outcomes and disease characteristics. Survival analysis using the Kaplan-Meier plotter database revealed that high p53 expression was significantly associated with worse OS and RFS in BRCA, CSCC, ESCC, LIHC, STAD, thymoma (THYM), UCEC, HNSC, KIRP, ovarian serous cystadenocarcinoma (OV), pheochromocytoma and paraganglioma (PCPG) and THCA (P < 0.05) (Fig. 3A, D). PD-1 expression correlated with prognosis in BLCA, BRCA, CSCC, HNSC, KIRC, KIRP, LIHC, OV, READ, SARC, STAD, TGCT and UCEC (P < 0.05) (Fig. 3B, E). PD-L1 showed significant associations with survival outcomes in BLCA, BRCA, CSCC, HNSC, KIRC, LIHC, LUAD, LUSC, OV, PAAD (pancreatic adenocarcinoma), SARC, THCA and UCEC (P < 0.05) (Fig. 3C, F). Further analysis using the UALCAN database specifically for BLCA revealed that low PD-1 expression was associated with poorer prognosis (Fig. 3H), while p53 and PD-L1 expression had no effect on BLCA outcomes (Fig. 3G, I). These results suggest that p53 and PD-1/PD-L1 expression significantly influence prognosis in multiple cancers, particularly BRCA, CSCC, HNSC, LIHC, OV, and UCEC.

Fig. 3 [Images not available. See PDF.]

Evaluated the survival curve of p53 and PD-1/PD-L1 for the prognostic value of pan-cancer. (A, D) Forest plots show the correlation between p53 expression and OS and RFS of pan-cancer. (B, E) Correlation between PD-1 expression and OS and RFS of pan-cancer. (C, F) Correlation between PD-L1 expression and OS and RFS of pan-cancer. (G-I) Prognostic value and differential p53 and PD-1/PD-L1 expression in BLCA patients was examined by using UALCAN database.

In addition, clinicopathological correlation studies in UC using UALCAN database demonstrated that p53 and PD-L1 expression varied significantly with individual cancer stage (2,3 and 4), gender (male and female), age (41–60 and 61–80 years), body weight (normal weight, extreme weight, and obese), tumor histology (papillary and non-papillary), molecular subtypes (neuronal, basal squamous, luminal, and luminal papillary), and lymph node metastatic status (N0 and N2) (Fig. 4A, C). PD-1 expression showed significant variation only across molecular subtypes (Fig. 4B). At the same time, the correlation between p53 and PD-1/PD-L1 protein expression and clinicopathological features of UC were also assessed in our clinical data. The results showed that p53 protein overexpression was significantly associated with tumor diameter, tumor multiplicity, tumor grade, tumor stage, and lymph node metastasis. PD-L1 protein overexpression was significantly associated with tumor diameter, tumor grade, and tumor stage (Table 3), while PD-1 only showed association with tumor grade. Therefore, these results suggest that p53 and PD-1/PD-L1 may be used as biomarkers to assess the condition of UC patients.

Fig. 4 [Images not available. See PDF.]

Box plots evaluating the associations between p53 and PD-1/PD-L1 mRNA expression and different clinicopathological parameters using the UALCAN database. (A-C) The correlation between p53 and PD-1/PD-L1 expression and UC clinicopathological parameters, including individual cancer stages, gender, weight, age, tumor histology, molecular subtype, and nodal metastasis status. *P < 0.05, **P < 0.01, ***P < 0.001.

Table 3. The clinicopathologic significance of p53 and PD-1/PD-L1 expression in patients with UC.

Characteristics	Case No.	p53		P	PD-1		P	PD-L1		P
		High	Low		High	Low		High	Low
Age, years
< 65	32	22	10	0.857	10	22	0.934	14	18	0.914
≥ 65	33	22	11		10	23		14	19
Gender
Male	43	30	13	0.617	12	31	0.485	19	24	0.801
Female	22	14	8		8	14		9	13
Tumor diameter, cm
< 3	35	28	7	0.022	11	24	0.901	20	15	0.013
≥ 3	30	16	14		9	21		8	22
Tumor multiplicity
Unifocal	28	24	4	0.015	7	21	0.545	11	17	0.591
Multifocal	37	20	17		13	24		17	20
Tumor grade
Low	25	10	15	< 0.001	13	12	0.003	15	10	0.029
High	40	34	6		7	33		13	27
Tumor stage
Ta-T1	18	8	10	0.013	7	11	0.380	3	15	0.008
T2-T4	47	36	11		13	34		25	22
Nodal metastasis
Absent	54	40	14	0.030	14	40	0.061	21	33	0.184
Present	11	4	7		6	5		7	4
Distant metastasis
Absent	56	39	17	0.649	18	38	0.834	25	31	0.785
Present	9	5	4		2	7		3	6

Immune correlation and infiltration patterns of p53 and PD-1/PD-L1 in UC

To comprehensively investigate the molecular relationships between p53 and PD-1/PD-L1 and their roles in shaping the tumor immune microenvironment in UC. GEPIA2 database was used to revealed important but nuanced relationships between p53 and PD-1/PD-L1 mRNA expression in UC. In adjacent normal tissues, we found no significant correlations between these markers (Fig. 5A). However, there was a weak but statistically significant negative correlation between p53 and PD-1 expression (r=−0.12, P = 0.018), a strong positive correlation between PD-1 and PD-L1 expression (r = 0.65, P = 3.4e-50), while no correlation between p53 and PD-L1 expression in UC (r = 0.048, P = 0.33) (Fig. 5B). Next, we analyzed the correlation between the protein expressions of p53 and PD-1/PD-L1 in our data and obtained similar results. The results showed, a negative correlation between p53 and PD-1 expression (r=−0.252, P = 0.043), a positive correlation between PD-1 and PD-L1 expression (r = 0.295, P = 0.017). Notably, the expression of p53 and PD-L1 was not significantly correlated in UC (r=−0.063, P = 0.616) (Table 4). In addition, the expression of p53 and PD-1/PD-L1 was also not significantly correlated in adjacent normal tissues (Supplementary table S3). Collectively, these results indicated a functionally relevant association between p53 and PD-1, laying crucial groundwork for subsequent research. While the correlation strength was week, these preliminary findings merit deeper exploration to uncover the precise molecular interplay.

Fig. 5 [Images not available. See PDF.]

The correlation between the mRNA expression of p53 and PD-1/PD-L1 in BLCA tissues and adjacent normal tissues. (A) Correlation between the expression of p53 and PD-1/PD-L1 in adjacent normal tissues. (B) Correlation between the expression of p53 and PD-1/PD-L1 in UC tissues.

Table 4. The correlation between the expression of p53 and PD-1/PD-L1 in UC cancer tissues.

Groups	Total	PD-1		r	P	PD-L1		r	P
		High	Low			High	Low
p53
High	44	10	34	−0.252	0.043	18	26	−0.063	0.616
Low	21	10	11			10	11
PD-1
High	20	-	-			13	7	0.295	0.017
Low	45	-	-			15	30

GeneMANIA and STRING network analyses identified the top 20 genes co-expressed with p53 and PD-1/PD-L1, including PDCD1LG2, MDM2, MYB, LCK, COP1, SIN3A, CD3E, TP53BP1, TP53BP2, CD3D, FBXO38, MDM4, TP63, CD80, CD3G, CMTM6, MAPKAPK5, PMAIP1, DECR1, and TP73 (Fig. 6A-B). These genes demonstrated direct and indirect relationships between p53 and PD-1/PD-L1. Subsequently, miRNA prediction was performed on 20 key genes to construct a ceRNA network consisting of 10 mRNAs, 37 miRNAs, and 158 lncRNAs (Fig. 6C). This network architecture further substantiates the potential intrinsic regulatory relationships between p53 and PD-1/PD-L1. GO enrichment analysis revealed these genes participate in adaptive immune responses, T cell costimulation, alpha-beta T cell receptor complex formation, and T cell receptor binding (Fig. 6D). KEGG pathway analysis showed involvement in T-cell receptor signaling, Th1/Th2 cell differentiation, and Th17 cell differentiation pathways (Fig. 6E).

Fig. 6 [Images not available. See PDF.]

Genes and pathways closely related to p53 and PD-1/PD-L1. (A) The gene-gene interaction network of p53 and PD-1/PD-L1 were constructed using GeneMANIA database. (B) The PPI network of p53 and PD-1/PD-L1 were generated using STRING. (C)The ceRNA network for the key genes. (D) Significantly GO and (E) KEGG enrichment analysis of p53 and PD-1/PD-L1. The use of KEGG images was conducted under the guidance of the KEGG database (https://www.kegg.jp/) and with permission from the Kanehisa Laboratory (Ref: 251924).

Besed on the results of GO and KEGG, we performed comprehensive analysis of immune cell infiltration Using the TIMER database in BLCA. The results showed p53 expression positively correlated with tumor purity (r = 0.137, P = 8.35e-03) but negatively with CD8 + T cells (r=−0.143, P = 6.27e-03), neutrophils (r=−0.11, P = 3.63e-02) and DCs (r=−0.148, P = 4.58e-03) (Fig. 7A). In contrast, PD-1 showed positive correlations with CD8 + T cells (r = 0.251, P = 1.15e-06), CD4 + T cells (r = 0.384, P = 3.01e-14), neutrophils (r = 0.602, P = 3.27e-37) and DCs (r = 0.594, P = 3.80e-36), while negatively correlating with tumor purity (r=−0.554, P = 4.83e-31) (Fig. 7B). PD-L1 exhibited similar positive associations with CD8 + T cells (r = 0.422, P = 3.03e-17), CD4 + T cells (r = 0.211, P = 4.75e-05), neutrophils (r = 0.622, P = 3.44e-40) and DCs (r = 0.643, P = 5.09e-44), along with negative correlation with tumor purity (r=−0.397, P = 2.42e-15) (Fig. 7C). Then, We further investigated how somatic copy number alterations (SCNA) of these genes affected immune infiltration. The CNA of p53 significantly correlated with infiltration levels of CD4 + T cells, macrophages, neutrophils and DCs (Fig. 7D); PD-1 CNA with CD4 + T cells, neutrophils and DCs (Fig. 7E); and PD-L1 CNA with CD8 + T cells, CD4 + T cells, neutrophils and DCs (Fig. 7F). Moreover, immune subtype analysis in the TISIDB database showed p53 was highly expressed in C4 (lymphocyte depleted) and C3 (inflammatory) subgroups but low in C6 (TGF-β dominant) (Fig. 7G); PD-1 was high in C2 (IFN gamma dominant) and low in C6 (Fig. 7H); PD-L1 was high in C2 but low in C4 (Fig. 7I). These findings suggest that these markers can substantially reshape the tumor immune landscape.

Fig. 7 [Images not available. See PDF.]

Relationship between p53 and PD-1/PD-L1 expression and immune infiltration level of BLCA, CNV and immune subtypes. (A-C) Correlation between p53 and PD-1/PD-L1 expression and different type of immune cells in BLCA. (D) p53 and PD-1/PD-L1 CNV affects the infiltrating levels of CD4 + T cells, macrophages, neutrophils, and DCs in BLCA. (E) Distribution of p53 and PD-1/PD-L1 expression across immune subtypes in BLCA (TISIDB). (C1: wound healing; C2: IFN-gamma dominant; C3: inflammatory; C4: lymphocyte-depleted and C6: TGF-β dominant).

Notably, detailed immune marker correlation analysis (adjusted for tumor purity) demonstrated p53 was negative associations with TAM markers (CCL2, CD68, IL10), monocyte markers (CD86, CSF1R), M2 macrophage markers (CD163, VSIG4, MS4A4A), and DC markers (HLA-DQB1, HLA-DRA, HLA-DPA1, NRP1, ITGAX). Similarly, PD-1 and PD-L1 expression were significantly positively associated with TAM (CCL2, CD68, IL10), monocytes (CSF1R), M2 macrophages (CD163, VSIG4, MS4A4A) and DCs (HLA-DPB1, HLA-DQB1, HLA-DRA, HLA-DPA1, NRP1, and ITGAX) (Table 5). Thus, these results indicate that p53 and PD-1/PD-L1 expression are significantly associated with immune infiltration in the tumor microenvironment of UC and play an important role in its immune escape.

Table 5. Correlation analysis between p53 and PD-1/PD-L1 and biomarkers of immune cells in UC.

Description	Gene marker	TP53				PDCD1				CD274
		None		Purity		None		Purity		None		Purity
		Cor	P	Cor	P	Cor	P	Cor	P	Cor	P	Cor	P
B cell	CD19	−0.102	*	−0.054	0.304	0.532	***	0.362	***	0.246	***	0.076	0.144
	CD79A	−0.123	*	−0.072	0.168	0.657	***	0.502	***	0.321	***	0.126	*
CD8 + T cell	CD8A	−0.197	***	−0.150	**	0.904	***	0.868	***	0.640	***	0.542	***
	CD8B	−0.106	*	−0.06	0.250	0.747	***	0.702	***	0.455	***	0.339	***
CD4 + T cell	CD4	−0.103	*	−0.031	0.548	0.784	***	0.684	***	0.511	***	0.356	***
TAM	CCL2	−0.144	**	−0.110	*	0.576	***	0.380	***	0.461	***	0.297	***
	CD68	−0.133	**	−0.079	0.128	0.574	***	0.442	***	0.472	***	0.352	***
	IL10	−0.159	**	−0.123	*	0.620	***	0.446	***	0.477	***	0.331	***
T cell (general)	CD3D	−0.142	**	−0.083	0.112	0.876	***	0.809	***	0.515	***	0.360	***
	CD3E	−0.171	***	−0.112	*	0.926	***	0.890	***	0.582	***	0.449	***
	CD2	−0.180	***	−0.124	*	0.935	***	0.902	***	0.597	***	0.478	***
Monocyte	CD86	−0.210	***	−0.172	***	0.780	***	0.668	***	0.617	***	0.506	***
	CSF1R	−0.163	***	−0.104	*	0.747	***	0.607	***	0.560	***	0.426	***
M1 Macrophage	NOS2	0.033	0.503	0.064	0.223	0.250	***	0.181	***	0.252	***	0.209	***
	IRF5	−0.000	0.997	−0.009	0.869	−0.012	0.801	0.006	0.910	−0.070	0.160	−0.063	0.231
	PTGS2	−0.037	0.459	−0.025	0.631	0.091	0.067	−0.009	0.871	0.238	***	0.177	***
M2 Macrophage	CD163	−0.198	***	−0.148	**	0.742	***	0.601	***	0.569	***	0.437	***
	VSIG4	−0.180	***	−0.130	*	0.745	***	0.618	***	0.551	***	0.422	***
	MS4A4A	−0.206	***	−0.163	**	0.756	***	0.627	***	0.531	***	0.389	***
Neutrophil	CEACAM8	−0.023	0.637	−0.041	0.433	−0.082	0.099	−0.053	0.308	−0.088	0.077	−0.087	0.095
	ITGAM	−0.151	**	−0.113	*	0.715	***	0.569	***	0.568	***	0.449	***
	CCR7	0.047	0.340	0.065	0.216	0.128	**	0.020	0.698	−0.041	0.409	−0.132	*
Dendritic cell	HLA-DPB1	−0.203	***	−0.153	**	0.838	***	0.760	***	0.556	***	0.429	***
	HLA-DQB1	−0.223	***	−0.170	**	0.774	***	0.673	***	0.545	***	0.429	***
	HLA-DRA	−0.182	***	−0.121	*	0.821	***	0.738	***	0.613	***	0.512	***
	HLA-DPA1	−0.196	***	−0.142	*	0.813	***	0.732	***	0.590	***	0.486	***
	CD1C	−0.064	0.199	−0.013	0.811	0.378	***	0.219	***	0.197	***	0.058	0.270
	NRP1	−0.101	*	−0.047	0.372	0.407	***	0.258	***	0.499	***	0.398	***
	ITGAX	−0.212	***	−0.171	***	0.759	***	0.630	***	0.578	***	0.444	***
Th1	TBX21	−0.185	***	−0.146	**	0.891	***	0.839	***	0.597	***	0.476	***
	STAT4	−0.210	***	−0.156	**	0.762	***	0.651	***	0.617	***	0.506	***
	STAT1	−0.161	**	−0.126	*	0.694	***	0.612	***	0.728	***	0.684	***
	IFNG	−0.222	***	−0.190	***	0.832	***	0.800	***	0.633	***	0.574	***
	TNF	0.024	0.623	0.070	0.177	0.447	***	0.311	***	0.559	***	0.464	***
Th2	GATA3	0.134	**	0.066	0.209	−0.363	***	−0.230	***	−0.408	***	−0.318	***
	STAT6	0.138	**	0.116	*	−0.169	*	−0.025	0.638	−0.141	**	−0.104	*
	STAT5A	−0.009	0.852	0.018	0.731	0.444	***	0.324	***	0.288	***	0.148	**
	IL13	−0.039	0.435	−0.006	0.915	0.368	***	0.282	***	0.262	***	0.176	***
Tfh	BCL6	0.037	0.460	0.026	0.622	−0.135	**	−0.061	0.246	−0.228	***	−0.200	***
	IL21	0.004	0.935	0.041	0.429	0.396	***	0.345	***	0.354	*	0.300	***
Th17	STAT3	−0.085	0.086	−0.053	0.306	0.383	***	0.266	***	0.503	***	0.432	***
	IL17A	0.058	0.240	0.057	0.275	0.229	***	0.232	***	0.111	*	0.090	0.083
Treg	FOXP3	−0.143	**	−0.091	0.082	0.779	***	0.670	***	0.599	***	0.488	***
	CCR8	−0.140	**	−0.084	0.107	0.653	***	0.506	***	0.578	***	0.478	***
	STAT5B	0.035	0.478	0.013	0.803	0.066	0.181	0.078	0.137	0.075	0.130	0.076	0.145
	TGFB1	−0.083	0.095	−0.024	0.649	0.298	***	0.190	***	0.325	***	0.255	***
T cell exhaustion	HAVCR2	−0.219	***	−0.190	***	0.851	***	0.777	***	0.647	***	0.549	***
	LAG3	−0.191	***	−0.157	**	0.875	***	0.820	***	0.681	***	0.596	***
	CXCL13	−0.161	**	−0.122	*	0.810	***	0.741	***	0.566	***	0.452	***
	LAYN	−0.134	**	−0.086	0.098	0.441	***	0.235	***	0.332	***	0.158	**

*P<0.05; **P<0.01; ***P<0.001.

The survival analysis revealed complex, context-dependent prognostic associations that varied significantly based on specific immune cell infiltration patterns. Our results indicate that high expression of p53 in decreased CD4 + memory T cells, CD8 + T cells, and Th2 cells, as well as low expression in enriched natural killer T cells and Th2 cells, can lead to poorer OS in BLCA patients. However, high expression of p53 in decreased CD4 + memory T cells, CD8 + T cells, mesenchymal stem cells, and low expression in enriched CD8 + cells and decreased Th2 cells can lead to poorer RFS in BLCA patients. However, there was no significant correlation between the expression of p53 and the prognosis of BLCA in different B cells, macrophages, regulatory T cells, and Th1 cells (Fig. 8A-B). In addition, low expression of PD-1 in enriched B cells, CD4 + memory T cells, CD8 + T cells, macrophages, mesenchymal stem cells, natural killer T cells, Th1 cells, as well as decreased CD8 + T cells and natural killer T cells can lead to poor OS in BLCA patients (Fig. 8C). However, the expression of PD-1 only in decreased B cells and enriched macrophages was not significantly correlated with the RFS of BLCA (Fig. 8D). Meanwhile, we found that low expression of PD-L1 in enriched B cells, macrophages, mesenchymal stem cells, natural killer T cells, and Th1 cells, as well as high expression in reduced CD4 + memory T cells and Th1 cells, resulted in poor OS in BLCA patients (Fig. 8E). BLCA patients with low expression of PD-L1 showed poorer RFS in enriched CD8 + T cells and macrophages, as well as decreased mesenchymal stem cells, NK cells, and TH2 cells (Fig. 8F). These results indicate that the differential expression of p53 and PD-1/PD-L1 can affect the prognosis of BLCA patients, partially attributed to tumor cell immune infiltration.

Fig. 8 [Images not available. See PDF.]

Prognostic evaluation of p53 and PD-1/PD-L1 expression based on immune cells in BLCA patients. Forest plots showing the correlations between (A) OS and (B) RFS and the p53 expression according to different immune cell subgroups. Correlations between PD-1 expression and OS (C) and RFS (D) in different immune cell subgroups. Correlations between PD-L1 expression and OS (E) and RFS (F) in different immune cell subgroups.

Predictive value in immunotherapy based on p53 and PD-1/PD-L1 expression

To assess clinical utility, we evaluated these markers as predictors of immunotherapy response. Comprehensive analysis of TCGA-BLCA data revealed p53 expression negatively correlated with multiple immune checkpoints including CD44, CD86, CTLA4, HAVCR2, IDO1, PDCD1LG2 and TNFRSF18, while positively correlating only with TMIGD2 (Fig. 9A). PD-1 showed positive correlations with 38 checkpoints (e.g., ADORA2A, BTLA, CD160, CD200, CD200R1, CD244, CD27, CD274, CD276, CD28, CD40, CD40LG, CD44, CD48, CD70, CD80, CD86, CTLA4) and negative correlations with TNFSF15 and VTCN1 (Fig. 9A). PD-L1 positively correlated with 35 checkpoints and negatively with TNFSF15 (Fig. 9A). In addition, TMB is also an important indicator that affects the efficacy of immunotherapy. The results showed a significant negative correlation between p53 expression and TMB (r=−0.15, P = 0.047), and a significant positive correlation between PD-1/PD-L1 expression and TMB (r = 0.11, P = 0.029; r = 0.12, P = 0.023) (Fig. 9B). To evaluate BLCA immune escape and immune therapy response in BLCA, we calculated the TIDE score for each tumor sample, where higher TIDE scores indicate a higher likelihood of immune escape. The obtained results indicate that the TIDE scores of the PD-1 and PD-L1 high expression groups are significantly higher than those of the low expression group (Fig. 9D-E). Although no significant correlation was observed between p53 expression and TIDE score, the p53 low expression group showed a trend towards high TIDE score, which has certain reference value for predicting BLCA immunotherapy (Fig. 9C). Furthermore, the impact of p53, PD-1/PD-L1 expression on the efficacy of immunotherapy was evaluated in BLCA patients with different states of CTLA4 and PD-1. In BLCA patients with CTLA4 (-) PD-1 (-) and CTLA4 (+) PD-1 (-), patients with high p53 expression had IPS scores, indicating that these patients may benefit more from CTLA4-targeted treatments (Fig. 9F). However, when CTLA4 and PD-1 are arbitrarily highly expressed (+/+, +/- or -/+), patients in the PD-1/PD-L1 high expression group exhibit higher IPS scores (Fig. 9G-H). These findings suggest that p53 and PD-1/PD-L1 expression may effectively predict the risk of immune evasion and the response to immunotherapy in BLCA individuals.

Fig. 9 [Images not available. See PDF.]

Effect of p53, PD-1/PD-L1 expression on BLCA immunotherapy. (A) The correlation between p53, PD-1/PD-L1 expression and immune therapy checkpoints; (B) The correlation between the expression of p53, PD-1/PD-L1 and tumor mutation burden; (C) Comparison of TIDE scores between p53 high expression group and low expression group. (D) Comparison of TIDE scores between PD-1 high expression group and low expression group. (E) Comparison of TIDE scores between PD-L1 high expression group and low expression group. (F) The effect of p53 expression on the efficacy of immunotherapy in BLCA patients; (G) The effect of PD-1 expression on the efficacy of immunotherapy in BLCA patients; (H) The effect of PD-L1 expression on the efficacy of immunotherapy in BLCA patients.

Discussion

UC is the most common cancer in the urinary system. So far, immunotherapy based on PD-1/PD-L1 has been widely applied to UC patients and has achieved some success, but there are still many UC patients with low reactivity to it^33,34. Therefore, exploring the relevant molecular mechanisms of PD-1/PD-L1 in UC and identifying an immune-related molecular marker is crucial for improving immunotherapy response and prognosis. Although the involvement of p53 and PD-1/PD-L1 in the growth and prognosis of UC has been partially confirmed^{35, 36–37}, the correlation, specific molecular alterations, and biological functions of these genes or proteins are not fully understood. Therefore, our study further explored the roles and correlations of these genes and proteins in UC, especially in its immune microenvironment.

To detect the expression and prognosis of p53 and PD-1/PD-L1 in pan-cancer, we first conducted extensive data retrieval from multiple publicly available databases. The results showed that the mRNA expression of p53 and PD-1/PD-L1 was significantly upregulated in various cancer tissues compared to normal tissues. Subsequently, through Kaplan-Meier survival analysis, p53 and PD-1/PD-L1 were identified as one of the factors contributing to poor prognosis in various human cancers. At the same time, it was found that PD-1 expression was significantly associated with poorer OS and RFS in BLCA patients, while p53 and PD-L1 expression were not associated with the prognosis of BLCA. Finally, the significant correlation between the low expression of PD-1 and poor prognosis in BLCA was further demonstrated by the UALCAN database. These results indicate that p53 and PD-1/PD-L1 are prognostic markers for various cancers, while PD-1 is only a prognostic marker for BLCA.

To investigate the expression patterns of p53 and PD-1/PD-L1 in urothelial carcinoma (UC) and their clinical relevance, we initially analyzed data from the UALCAN and GEPIA2 databases. The results revealed significantly elevated mRNA expression levels of p53 and PD-L1 in UC tissues compared to normal controls, while PD-1 expression showed no significant difference between UC and normal tissues. Meanwhile, protein level validation using clinical samples showed a significant increase in the expression of p53 and PD-L1 in BLCA and UC cancer tissues compared to adjacent normal tissues, while there was no significant difference in the expression of PD-1. Unfortunately, p53 and PD-1/PD-L1 expression did not significantly differ between UTUC and adjacent normal tissues. We speculate that this may be caused by a small sample size. Subsequently, it was demonstrated through the UALCAN database that the expression of p53 and PD-L1 was associated with multiple clinicopathological parameters, while PD-1 was only associated with molecular subtypes. Similarly, clinical data was applied for validation, and it was found that p53 and PD-L1 were associated with tumor diameter, tumor grade, and tumor stage, while PD-1 was only associated with tumor grade. In summary, these results indicated that p53 and PD-1/PD-L1 are biomarkers for the growth and progression of UC.

In order to further explore the functions and mechanisms of p53 and PD-1/PD-L1 in UC, the correlation between p53 and PD-1/PD-L1 was first explored through the GEPIA2 database. The results showed that p53 was significantly negatively correlated with PD-1, PD-1 was significantly positively correlated with PD-L1, but p53 was not significantly correlated with PD-L1. Again, validation by applying clinical samples showed similar results. Next, we explored the co-expression genes and proteins of p53 and PD-1/PD-L1 through the GeneMANIA, STRING and ceRNA network, and enriched and analyzed these genes using GO and KEGG. The results showed that the biological processes and KEGG pathways associated with these genes include adaptive immune responses, T cell costimulation, alpha-beta T cell receptor complex, T cell receptor binding, T-cell receptor signaling pathways, Th1 and Th2 cell differentiation, Th17 cell differentiation. These results indicated that p53 and PD-1/PD-L1 may affect the growth and prognosis of UC by regulating its tumor microenvironment. Furthermore, considering the correlation between p53 and PD-1/PD-L1, we speculated that upregulation of p53 expression suppresses the expression of PD-1, thereby suppressing the PD-1/PD-L1 axis to reduce immunosuppressive status and reactivate immune cells to kill tumor cells. Therefore, we further investigated the relationship between p53, PD-1/PD-L1 and different levels of immune cell infiltration in the tumor microenvironment.

The tumor microenvironment is composed of various immune and inflammatory cells, which play a crucial role in tumorigenesis and progression, and have significant implications for chemotherapy resistance and immunotherapy^38,39. Therefore, studying the tumor immune microenvironment and identifying potential immunotherapy targets are key choices for improving the effectiveness of patient immunotherapy. In this study, the relationship between the expression levels of p53, PD-1/PD-L1 and the level of immune infiltration of different immune cells was investigated to some extent. Our results indicated that the expression level of p53 was significantly negatively correlated with the infiltration level of CD8 + T cells, neutrophils, and DCs; The expression levels of PD-1 and PD-L1 were significantly positively correlated with the infiltration levels of CD8 + T cells, CD4 + T cells, neutrophils, and DCs. Consistently, it was found that there was a significant negative correlation between p53 expression and gene markers of CD8 + T cells, TAMs, monocytes, M2 macrophages, and DCs, while PD-1 and PD-L1 expression were significantly positively correlated with gene markers of TAMs, monocytes, M2 macrophages, and DCs.

Currently, the relationship between TAMs polarization and PD-1/PD-L1 axis is complex and has not been thoroughly elucidated. Generally speaking, there are two subtypes of TAMs, including M1 macrophages (classically activated) and M2 macrophages (alternatively activated)⁴⁰. M1 macrophages induce anti-tumor immune responses through their T cell stimulating activity, while M2 macrophages promote tumor immune escape, angiogenesis, tumor growth, and metastasis^{41, 42–43}. Previous studies have shown that TAMs can be tilted by PD-1/PD-L1 from M1 to M2 phenotype, and M2 TAMs have been found to promote resistance to PD-1/PD-L1 targeted drug^44,45. In addition, TAMs can upregulate the expression of PD-1/PD-L1 in tumor cells to mediate tumor immune escape through various pathways^46,47. Therefore, TAMs are considered key targets for immunotherapy. In our study, the expression of p53 was significantly negatively correlated with the gene markers of TAMs and M2 macrophages, while the expression of PD-1 and PD-L1 was significantly positively correlated. This means that high expression of p53 may inhibit the polarization of TAMs and M2 macrophages, which may block the formation of PD-1/PD-L1 drug resistance by inducing M2/M1 transformation of TAMs. Secondly, in order to further explore the importance of p53 and PD-1/PD-L1 in immunotherapy, we explored the impact of the relationship between p53, PD-1/PD-L1 expression and immune cell infiltration on the prognosis and survival of BLCA patients. The results revealed that differential expression of p53 and PD-1/PD-L1 could affect the prognosis of BLCA patients, which was partially attributed to immune infiltration of tumor cells. Finally, it was further revealed that p53 and PD-1/PD-L1 had an important role in BLCA immunotherapy. TIDE analysis further supported the potential role of p53 and PD-1/PD-L1 in immune escape and immunotherapy of BLCA. Notably, p53 high expression achieved better immunotherapeutic effects in PD-1-negative BLCA patients, which again validated our previous speculation. In summary, p53 and PD-1/PD-L1 play important roles in regulating immune cell infiltration in UC. p53 may be a key target in regulating the PD-1/PD-L1 axis and may modulate macrophage polarization to block the formation of PD-1/PD-L1 resistance, which may become the key to immunotherapy breakthrough.

While our study provides novel insights into the interplay between p53 and PD-1/PD-L1 in UC, several important implications emerge from these findings. First, the observed correlation between p53 status and PD-1/PD-L1 expression patterns suggests that p53 may serve as a potential biomarker for predicting ICIs responsiveness in UC patients. This finding could have significant clinical implications, as it may help stratify patients who are more likely to benefit from PD-1/PD-L1-targeted therapies. Furthermore, our data indicating that high p53 expression may inhibit TAM polarization and M2 macrophage differentiation propose a new mechanism by which p53 status could shape the tumor immune microenvironment. However, several limitations must be acknowledged. The reliance on public database analyses, while providing comprehensive genomic data, lacks validation in dedicated immunotherapy cohorts. This limitation particularly affects our predictive models of immunotherapy response, which require prospective clinical validation. Secondly, while we identified significant associations at the mRNA level, the molecular mechanisms underlying p53’s potential regulation of the PD-1/PD-L1 axis in UC pathogenesis - including its roles in tumor progression, metastasis, and immune evasion - remain to be experimentally verified. Third, the protein-level dynamics of these interactions were not assessed in this study, representing an important gap given the potential post-transcriptional regulation of these pathways. These limitations directly point to critical future research directions: (1) functional studies to mechanistically validate p53’s regulatory role in PD-1/PD-L1 expression and TAM polarization; (2) proteomic analyses to complement our transcriptomic findings; (3) prospective clinical studies correlating p53 status with actual ICI treatment outcomes. Most importantly, our findings suggest that therapeutic strategies simultaneously targeting both p53 and PD-1/PD-L1 pathways may represent a promising approach for UC treatment, warranting further preclinical investigation.

Therefore, to elucidate the negative regulatory relationship between p53 and PD-1 in BLCA, we propose a comprehensive research strategy integrating molecular, genomic, and in vivo approaches. First, we will establish isogenic BLCA cell lines with defined p53 status (wild-type, knockout, and over-expression) to systematically assess PD-1 expression dynamics at transcriptional, translational, and surface protein levels using RT-qPCR, Western blot, and flow cytometry. Concurrently, we will perform computational prediction of putative p53 binding sites in PD-1 regulatory regions, followed by experimental validation through chromatin immunoprecipitation sequencing (ChIP-seq) and luciferase reporter assays to determine whether p53 directly modulates PD-1 expression at the transcriptional level. To evaluate the functional consequences of this regulatory axis in an immunocompetent setting, we will develop orthotopic BLCA models using syngeneic MB49 cells with engineered p53/PD-1 status and patient-derived xenografts, which will be treated with anti-PD-1 therapy alone or in combination with p53-stabilizing agents. Tumor growth kinetics, immune cell infiltration patterns, and molecular markers will be comprehensively analyzed to determine how PD-1 activation influences p53-mediated tumor suppression. This multi-dimensional approach will provide mechanistic insights into p53-PD-1 crosstalk while establishing its clinical relevance for immunotherapy response prediction in BLCA patients.

Conclusions

In summary, p53, as an immune gene or protein associated with PD-1/PD-L1, was significantly negatively correlated with PD-1. High expression of p53 may inhibit PD-1 expression, further inhibiting the PD-1/PD-L1 axis to reduce immunosuppressive status. In addition, p53 may also block the formation of PD-1/PD-L1 resistance by inhibiting the polarization of TAMs and M2 macrophages. This work demonstrates the important role of p53 in PD-1/PD-L1 axis-based immunotherapy for UC patients, and p53 is expected to become a key target for breaking through the current status of immunotherapy.

Acknowledgements

Thanks all the research participants of this study.

Author contributions

Conceptualization, N.Z. and M.Z.; methodology, M.Z., and Y.L.; software, M.Z., and P.H.; validation, M.Z., J.T., and L.F.; formal analysis, T.H., P.H., and Y.L.; investigation, Y.Z., and L.F.; resources, J.T., T.H., and M.Z.; data curation, M.Z. and P.H.; visualization, M.Z. and Y.L.; supervision, N.Z.; project administration, N.Z. and X.L.; funding acquisition, N.Z. and X.L. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by National Natural Sciences Foundation of China (grant number 81860524) and the Science and Technology Department of Zunyi city (grant number HZ2020228).

Data availability

The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Publisher’s note

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

References

1. Siegel, RL; Giaquinto, AN; Jemal, A. Cancer statistics, 2024.CA. Cancer. J. Clin.; 2024; 74, pp. 12-49.

2. Alfred, WJ et al. European association of urology guidelines on Muscle-invasive and metastatic bladder cancer: summary of the 2023 guidelines. Eur. Urol.; 2024; 85, pp. 17-31. [DOI: https://dx.doi.org/10.1016/j.eururo.2023.08.016]

3. Compérat, E et al. Current best practice for bladder cancer: a narrative review of diagnostics and treatments. Lancet; 2022; 400, pp. 1712-1721. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36174585][DOI: https://dx.doi.org/10.1016/S0140-6736(22)01188-6]

4. Stone, L. Biomarkers for bladder cancer diagnosis, prognosis, monitoring and treatment. Nat. Rev. Urol.; 2022; 19, 383. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/35711060][DOI: https://dx.doi.org/10.1038/s41585-022-00620-2]

5. Birtle, A et al. Adjuvant chemotherapy in upper tract urothelial carcinoma (the POUT trial): a phase 3, open-label, randomized controlled trial. Lancet; 2020; 395, pp. 1268-1277.1:CAS:528:DC%2BB3cXkvVGgtbw%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32145825][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7181180][DOI: https://dx.doi.org/10.1016/S0140-6736(20)30415-3]

6. Rouprêt, M et al. European association of urology guidelines on upper urinary tract urothelial carcinoma: 2023 update. Eur. Urol.; 2023; 84, pp. 49-64. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36967359][DOI: https://dx.doi.org/10.1016/j.eururo.2023.03.013]

7. Zhao, J; Guo, CC; Rao, P. Contemporary issues in urothelial carcinoma of upper urinary tract. Adv. Anat. Pathol.; 2024; 31, pp. 80-87.1:CAS:528:DC%2BB2cXivFeisrw%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38009077][DOI: https://dx.doi.org/10.1097/PAP.0000000000000421]

8. Nadal, R; Valderrama, BP; Bellmunt, J. Progress in systemic therapy for advanced-stage urothelial carcinoma. Nat. Rev. Clin. Oncol.; 2024; 21, pp. 8-27.1:CAS:528:DC%2BB3sXitlGiu77P [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37945764][DOI: https://dx.doi.org/10.1038/s41571-023-00826-2]

9. Knowles, M; Dyrskjøt, L; Heath, EI; Bellmunt, J; Siefker-Radtke, AO. Metastatic urothelial carcinoma. Cancer Cell.; 2021; 39, pp. 583-585.1:CAS:528:DC%2BB3MXhtVynurjM [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33974855][DOI: https://dx.doi.org/10.1016/j.ccell.2021.04.012]

10. Park, JH; Zhuang, J; Li, J; Hwang, PM. p53 as guardian of the mitochondrial genome. FEBS Lett.; 2016; 590, pp. 924-934.1:CAS:528:DC%2BC28Xht1KrtbjE [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/26780878][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4833664][DOI: https://dx.doi.org/10.1002/1873-3468.12061]

11. Luo, Q; Beaver, JM; Liu, Y; Zhang, Z. Dynamics of p53: A master decider of cell fate. Genes (Basel); 2017; 8, 66. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/28208785][DOI: https://dx.doi.org/10.3390/genes8020066]

12. Guo, G; Yu, M; Xiao, W; Cui, Y. Local activation of p53 in the tumor microenvironment overcomes immune suppression and enhances antitumor immunity. Cancer Res.; 2017; 77, pp. 2292-2305.1:CAS:528:DC%2BC2sXmvFart70%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/28280037][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5465961][DOI: https://dx.doi.org/10.1158/0008-5472.CAN-16-2832]

13. Morton, LM et al. Risk of non-Hodgkin lymphoma associated with germline variation in genes that regulate the cell cycle, apoptosis, and lymphocyte development. Cancer Epidemiol. Biomarkers Prev.; 2009; 18, pp. 1259-1270.1:CAS:528:DC%2BD1MXktFCgtr0%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/19336552][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2716008][DOI: https://dx.doi.org/10.1158/1055-9965.EPI-08-1037]

14. Dong, H et al. Tumor-associated B7-H1 promotes T-cell apoptosis: a potential mechanism of immune evasion. Nat. Med.; 2002; 8, pp. 793-800.1:CAS:528:DC%2BD38Xls12msLo%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/12091876][DOI: https://dx.doi.org/10.1038/nm730]

15. Grossman, HB et al. p53 and RB expression predict progression in T1 bladder cancer. Clin. Cancer Res.; 1998; 4, pp. 829-834.1:CAS:528:DyaK1cXisFels7o%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/9563875]

16. Krabbe, LM et al. Prognostic value of PD-1 and PD-L1 expression in patients with high grade upper tract urothelial carcinoma. J. Urol.; 2017; 198, pp. 1253-1262.1:CAS:528:DC%2BC2sXhsVOrsbvM [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/28668287][DOI: https://dx.doi.org/10.1016/j.juro.2017.06.086]

17. Ahadi, M et al. Evaluation of P53 and CK20 immunohistochemical markers in comparison with morphologic findings in Low- and High-grade urothelial carcinomas. Iran. J. Pathol.; 2021; 16, pp. 297-303. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34306125][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8298052][DOI: https://dx.doi.org/10.30699/ijp.2021.136330.2493]

18. Ku, JH et al. The role of p53 on survival of upper tract urothelial carcinoma: a systematic review and meta-analysis. Clin. Genitourin. Cancer; 2013; 11, pp. 221-228. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/23357513][DOI: https://dx.doi.org/10.1016/j.clgc.2012.12.009]

19. Tian, D et al. Expression of brain-specific angiogenesis inhibitor-1 and association with p53, microvessel density and vascular endothelial growth factor in the tissue of human bladder transitional cell carcinoma. Mol. Med. Rep.; 2015; 12, pp. 4522-4529.1:CAS:528:DC%2BC2MXhs12hsb3F [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/26129954][DOI: https://dx.doi.org/10.3892/mmr.2015.3984]

20. Kumar, U et al. Immunoexpression of PD-L1 and PD-1 and its clinicopathological correlation in urothelial carcinomas. J. Lab. Physicians; 2021; 14, pp. 197-201. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/35982880][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9381309]

21. Kim, B; Lee, C; Kim, YA; Moon, KC. PD-L1 expression in Muscle-Invasive urinary bladder urothelial carcinoma according to Basal/Squamous-Like phenotype. Front. Oncol.; 2020; 10, 527385. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33365265][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7750632][DOI: https://dx.doi.org/10.3389/fonc.2020.527385]

22. Shen, W et al. Sangerbox: A comprehensive, interaction-friendly clinical bioinformatics analysis platform. IMeta; 2022; 1, e36.1:CAS:528:DC%2BB2MXosFKrtr8%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38868713][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10989974][DOI: https://dx.doi.org/10.1002/imt2.36]

23. Chandrashekar, DS et al. UALCAN: A portal for facilitating tumor subgroup gene expression and survival analyses. Neoplasia; 2017; 19, pp. 649-658.1:CAS:528:DC%2BC2sXovVKhurw%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/28732212][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5516091][DOI: https://dx.doi.org/10.1016/j.neo.2017.05.002]

24. Tang, Z; Kang, B; Li, C; Chen, T; Zhang, Z. GEPIA2: an enhanced web server for large-scale expression profiling and interactive analysis. Nucleic Acids Res.; 2019; 47, pp. W556-W560.1:CAS:528:DC%2BB3cXisVCmur0%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31114875][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6602440][DOI: https://dx.doi.org/10.1093/nar/gkz430]

25. Warde-Farley, D et al. The genemania prediction server: biological network integration for gene prioritization and predicting gene function. Nucleic Acids Res.; 2010; 38, pp. W214-220.1:CAS:528:DC%2BC3cXotVSqtL4%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/20576703][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2896186][DOI: https://dx.doi.org/10.1093/nar/gkq537]

26. Franceschini, A et al. STRING v9.1: protein-protein interaction networks, with increased coverage and integration. Nucleic Acids Res.; 2013; 41, pp. D808-815.1:CAS:528:DC%2BC38XhvV2ksb3I [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/23203871][DOI: https://dx.doi.org/10.1093/nar/gks1094]

27. Ashburner, M et al. Gene ontology: tool for the unification of biology. The gene ontology consortium. Nat. Genet.; 2000; 25, pp. 25-29.1:CAS:528:DC%2BD3cXjtFSlsbc%3D [DOI: https://dx.doi.org/10.1038/75556] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/10802651][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3037419]

28. Kanehisa, M; Goto, S. KEGG: Kyoto encyclopedia of genes and genomes. Nucleic Acids Res.; 2000; 28, pp. 27-30.1:CAS:528:DC%2BD3cXhvVGqu74%3D [DOI: https://dx.doi.org/10.1093/nar/28.1.27] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/10592173][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC102409]

29. Kanehisa, M; Furumichi, M; Sato, Y; Matsuura, Y; Ishiguro-Watanabe, M. KEGG: biological systems database as a model of the real world. Nucleic Acids Res.; 2025; 53, pp. D672-D677. [DOI: https://dx.doi.org/10.1093/nar/gkae909] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/39417505]

30. Li, T et al. A web server for comprehensive analysis of tumor-infiltrating immune cells. Cancer Res.; 2017; 77, pp. e108-110.2017pota.book...L1:CAS:528:DC%2BC2sXhslOltbnE [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/29092952][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6042652][DOI: https://dx.doi.org/10.1158/0008-5472.CAN-17-0307]

31. Ru, B et al. TISIDB: an integrated repository portal for tumor-immune system interactions. Bioinformatics; 2019; 35, pp. 4200-4202.1:CAS:528:DC%2BB3cXhtVektL7M [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/30903160][DOI: https://dx.doi.org/10.1093/bioinformatics/btz210]

32. Nagy, Á; Munkácsy, G; Győrffy, B. Pancancer survival analysis of cancer hallmark genes. Sci. Rep.; 2021; 11, 6047.2021NatSR.11.6047N1:CAS:528:DC%2BB3MXmslKgsrs%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33723286][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7961001][DOI: https://dx.doi.org/10.1038/s41598-021-84787-5]

33. Sonpavde, G; Dranitsaris, G; Necchi, A. Improving the cost efficiency of PD-1/PD-L1 inhibitors for advanced urothelial carcinoma: A major role for precision medicine??. Eur. Urol.; 2018; 74, pp. 63-65. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/29653886][DOI: https://dx.doi.org/10.1016/j.eururo.2018.03.015]

34. de Lista, AG et al. Clinical outcome after progressing to frontline and second-line Anti-PD-1/PD-L1 in advanced urothelial cancer. Eur. Urol.; 2020; 77, pp. 269-276. [DOI: https://dx.doi.org/10.1016/j.eururo.2019.10.004]

35. Hodgson, A; Xu, B; Downes, MR. p53 immunohistochemistry in high-grade urothelial carcinoma of the bladder is prognostically significant. Histopathology; 2017; 71, pp. 296-304. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/28342221][DOI: https://dx.doi.org/10.1111/his.13225]

36. Chen, X et al. Correlation between PD-L1 expression of the tumour cells and lymphocytes infiltration in the invasive front of urothelial carcinoma. J. Clin. Pathol.; 2023; 77, pp. 61-67. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36319076][DOI: https://dx.doi.org/10.1136/jcp-2021-207795]

37. Ariafar, A et al. Clinical relevance and prognostic significance of PD-1/PD-Ls in non-metastatic bladder cancer: A role for PD-L2. Mol. Immunol.; 2020; 124, pp. 35-41.1:CAS:528:DC%2BB3cXhtV2ru77P [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32512320][DOI: https://dx.doi.org/10.1016/j.molimm.2020.05.010]

38. Hui, L; Chen, Y. Tumor microenvironment: sanctuary of the devil. Cancer Lett.; 2015; 368, pp. 7-13.1:CAS:528:DC%2BC2MXhtleksLfL [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/26276713][DOI: https://dx.doi.org/10.1016/j.canlet.2015.07.039]

39. Tang, T et al. Advantages of targeting the tumor immune microenvironment over blocking immune checkpoint in cancer immunotherapy. Signal. Transduct. Target. Ther.; 2021; 6, 72. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33608497][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7896069][DOI: https://dx.doi.org/10.1038/s41392-020-00449-4]

40. Namgaladze, D et al. Polarization of human macrophages by Interleukin-4 does not require ATP-citrate lyase. Front. Immunol.; 2018; 9, 2858.1:CAS:528:DC%2BC1MXht1eqt7bM [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/30568658][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6290342][DOI: https://dx.doi.org/10.3389/fimmu.2018.02858]

41. Martinez, FO; Gordon, S. The M1 and M2 paradigm of macrophage activation: time for reassessment. F1000Prime Rep.; 2014; 6, 13. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/24669294][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3944738][DOI: https://dx.doi.org/10.12703/P6-13]

42. Murray, PJ et al. Macrophage activation and polarization: nomenclature and experimental guidelines. Immunity; 2014; 41, pp. 14-20.1:CAS:528:DC%2BC2cXhtFOrsLjM [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/25035950][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4123412][DOI: https://dx.doi.org/10.1016/j.immuni.2014.06.008]

43. Ivashkiv, LB. Epigenetic regulation of macrophage polarization and function. Trends Immunol.; 2013; 34, pp. 216-223.1:CAS:528:DC%2BC38XhslymtL3O [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/23218730][DOI: https://dx.doi.org/10.1016/j.it.2012.11.001]

44. Wei, Y et al. PD-L1 induces macrophage polarization toward the M2 phenotype via erk/akt/mtor. Exp. Cell. Res.; 2021; 402, 112575.1:CAS:528:DC%2BB3MXot1Ghu7o%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33771483][DOI: https://dx.doi.org/10.1016/j.yexcr.2021.112575]

45. Roulleaux Dugage, M; Jones, RL; Trent, J; Champiat, S; Dumont, S. Beyond the driver mutation: immunotherapies in Gastrointestinal stromal tumors. Front. Immunol.; 2021; 12, 7157279. [DOI: https://dx.doi.org/10.3389/fimmu.2021.715727]

46. Zhang, H et al. Roles of tumor-associated macrophages in anti-PD-1/PD-L1 immunotherapy for solid cancers. Mol. Cancer; 2023; 22, 58. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36941614][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10029244][DOI: https://dx.doi.org/10.1186/s12943-023-01725-x]

47. Li, W et al. Correlation between PD-1/PD-L1 expression and polarization in tumor-associated macrophages: A key player in tumor immunotherapy. Cytokine Growth Factor. Rev.; 2022; 67, pp. 49-57.1:CAS:528:DC%2BB38XhvVOhsLvI [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/35871139][DOI: https://dx.doi.org/10.1016/j.cytogfr.2022.07.004]