Abstract

Background

   XPR1 is crucial in the development of some tumors, yet its association
   with endometrial cancer (EC) remains uncertain. We propose that XPR1
   exhibits elevated expression in EC and is significantly linked to
   unfavorable patient outcomes, positioning it as a prospective
   prognostic biomarker.

Methods

   This study investigated XPR1 expression in 554 Uterine Corpus
   Endometrial Carcinoma(UCEC) cases and 35 normal tissue samples using
   the The Cancer Genome Atlas(TCGA) database. Western blotting confirmed
   XPR1 protein presence in EC cell lines (ECC-1) and normal endometrial
   cells (EEC). The impact of XPR1 on EC cell proliferation and invasion
   was assessed using EdU proliferation and Transwell invasion assays. A
   Dot blot assay evaluated m6A methylation of XPR1 in ECC-1 and EEC. The
   relationship between XPR1 and EC patient prognosis was analyzed using
   Kaplan–Meier survival and Cox regression analyses. A nomogram was
   developed to predict 1-, 3-, and 5-year survival probabilities for EC
   patients, with its accuracy assessed via a calibration curve.
   Functional enrichment analyses using Gene Ontology (GO) and Kyoto
   Encyclopedia of Genes and Genomes (KEGG) were conducted to elucidate
   biological roles and pathways related to XPR1-associated genes. The
   'corrplot' R package analyzed the correlation between XPR1 expression
   and m6A methylation in EC, as well as its link to immune infiltration.
   Statistical analyses were performed using R software, with a bilateral
   p-value < 0.05 considered statistically significant.

Results

   This study revealed that XPR1 expression was significantly elevated in
   EC tissues compared to normal tissues (p < 0.001). Its expression
   correlated with clinical characteristics including patient age, BMI,
   clinical stage, histological grade, and tumor invasiveness, suggesting
   its potential as a prognostic marker.Kaplan–Meier analysis demonstrated
   that high XPR1 expression correlated with reduced overall survival
   (HR = 1.60, 95% CI: 1.06–2.42, p = 0.025). However, multivariable Cox
   regression analysis did not identify XPR1 as an independent prognostic
   factor.Functional experiments indicated that elevated XPR1 expression
   promoted proliferative and invasive capacities in endometrial carcinoma
   cells. XPR1 expression also associated with infiltration levels of
   immune cells (B cells, CD8 + T cells, CD4 + T cells, macrophages,
   neutrophils, and dendritic cells), suggesting potential involvement in
   tumor immune microenvironment regulation.Co-expressed genes with XPR1
   were enriched in key biological processes including RNA processing, DNA
   metabolic regulation, and KEGG signaling pathways. Furthermore, XPR1
   showed positive correlations with multiple m6A-related genes, and high
   XPR1 expression in EC coincided with elevated m6A methylation levels.
   Although these findings indicate that there is a correlation between
   XPR1 and m6A modification, it is uncertain that XPR1 can directly
   regulate m6A modification. The specific mechanism of XPR1 in m6A
   mediated post transcriptional regulation needs further study.

Conclusion

   This study identified a significant association between elevated XPR1
   expression and adverse clinical outcomes in endometrial carcinoma
   patients, with additional connections observed to m6A modification and
   tumor immune microenvironment dynamics. As a novel prognostic
   indicator, XPR1 demonstrated correlations with tumor aggressiveness
   features and biological processes—including cellular proliferation,
   invasion, and m6A/immune-related pathways—suggesting its potential
   biological relevance in endometrial carcinogenesis. However, its
   independent prognostic value remains undefined, and direct mechanistic
   roles in m6A-mediated post-transcriptional modulation warrant
   subsequent experimental validation.

Supplementary Information

   The online version contains supplementary material available at
   10.1186/s12885-025-14818-1.

   Keywords: XPR1, Endometrial cancer, M6A, Immune infiltration

Introduction

   Endometrial cancer (EC) is a major malignancy in the female
   reproductive system, significantly impacting patient health and
   healthcare systems globally. It ranks as the second most common
   gynecological cancer after cervical cancer, accounting for 20%−30% of
   such tumors [[30]1–[31]3]. In some developed cities, its incidence is
   the highest among gynecological malignancies. EC is classified into
   various subtypes based on histological and molecular characteristics,
   with endometrioid carcinoma being the most prevalent, comprising
   75%−80% [[32]4–[33]6]. This subtype is closely linked to estrogen
   exposure, generally well-differentiated, and has a favorable
   prognosis.In contrast, non-endometrioid subtypes like uterine serous
   carcinoma (USC) and clear cell carcinoma are less common but more
   aggressive with poorer prognoses [[34]7, [35]8]. TCGA project
   identifies four molecular subtypes of endometrial cancer, with serous
   cancer mainly falling under the'high copy number'category, marked by
   significant cell cycle imbalance and increased metastasis risk. These
   subtypes differ in epidemiology, molecular mechanisms, and clinical
   behavior. Understanding the molecular mechanisms of serous endometrial
   carcinoma and identifying reliable prognostic biomarkers is crucial for
   personalized treatment and improving patient outcomes. Research
   indicates that USC has distinct biological features, including high
   nuclear atypia and frequent TP53 gene mutations, underscoring the need
   to explore EC's molecular mechanisms and prognostic markers to inform
   clinical decisions [[36]9–[37]11].

   Xenotropic and polytropic retrovirus receptor 1 (XPR1) is a
   transmembrane protein integral to tumor development and
   progression,with elevated expression linked to poor prognosis in
   various cancers [[38]12–[39]14]. XPR1 functions as a phosphate efflux
   transporter, maintaining cellular phosphate homeostasis by mediating
   intracellular inorganic phosphorus (Pi) efflux [[40]15]. It influences
   tumor cell proliferation, invasion, and metastasis through modulation
   of phosphate metabolism and activation of signaling pathways. Notably,
   elevated XPR1 expression in endometrial carcinoma and other cancers
   correlates inversely with prognosis, suggesting its potential as a
   prognostic biomarker. RNA modification N6-methyladenosine (m6A)
   critically regulates tumorigenesis by influencing mRNA splicing,
   stability, transport, and translation.Analyses revealed statistically
   significant positive correlations (P < 0.05) between XPR1 expression in
   endometrial carcinoma tissues and key m6A regulators—including YTHDF1,
   YTHDF2, and METTL14—implying XPR1's potential association with m6A
   modification. Furthermore, immune infiltration assessments demonstrated
   significant inverse correlations (P < 0.05) between XPR1 and multiple
   immune cell populations, inferring XPR1's involvement in endometrial
   carcinoma immunoregulation.

   This study utilized bioinformatics analysis based on the TCGA and
   CBioPortal databases to systematically evaluate the relationship
   between XPR1 expression and clinical features and survival outcomes in
   EC patients. TCGA, a renowned public cancer research database, offers
   extensive gene expression and clinical follow-up data, enabling a
   thorough analysis of XPR1 expression patterns and their correlation
   with patient characteristics.CBioPortal enhances this analysis with its
   robust visualization and integration of multi-dimensional gene
   variation data, supporting investigations into XPR1 gene variation and
   related molecular mechanisms. These databases are authoritative and
   provide high-quality, large-sample data, ensuring the scientific rigor
   and reproducibility of the study's findings. The research focused on
   XPR1 as a novel molecular marker, integrating multi-omics data and
   functional experiments to explore its association with clinical
   characteristics and prognosis in EC patients.

   This study systematically evaluated the expression of the novel
   molecular marker XPR1 in EC by integrating multi-omics data and
   conducting functional experiments. It examined the association of XPR1
   with clinical characteristics and patient prognosis, exploring its
   potential in early screening, molecular typing, and precise prognosis
   assessment. The findings provide a new theoretical foundation and
   practical approach for clinical diagnosis and personalized management,
   surpassing traditional methods.

Materials and Methods

Data download and analysis

   mRNA expression data (TPM format) and clinical information for
   TCGA-UCEC were acquired from Xiantao Academic Online Analysis Tools
   (v2.1; [41]www.xiantaozi.com) on October 10, 2024, comprising 35 normal
   and 554 carcinoma tissue samples. Following quality control to exclude
   low-expression samples and outliers, subsequent analyses were conducted
   in the R environment (v4.2.1) integrated within the Xiantao platform.
   The packages ggplot2 (v3.4.4), stats (v4.2.1), and car (v3.1–0) were
   employed for pan-cancer analysis and differential expression
   visualization.The XPR1 gene's amplification and mutation status were
   sourced from cBioPortal, while the Human Protein Atlas (HPA) provided
   insights into the subcellular localization and immunofluorescence
   imaging of XPR1 in UCEC cells.

survival analysis

   Survival analysis assessing XPR1 expression impact on EC patient
   prognosis was performed using TCGA data. The analytical cohort was
   curated through sequential exclusion of normal tissue samples and
   clinically unannotated cases. Kaplan–Meier curves were generated using
   the survival package (v3.3.1) in R (v4.2.1) with patients dichotomized
   into high-/low-expression groups based on the 50% median expression
   threshold, and all results were visualized with ggplot2.

Correlation study of clinicopathological features

   Associations between XPR1 expression and clinical characteristics were
   analyzed using the Xiantao Academic online platform (R v4.2.1) with the
   packages ggplot2 (v3.4.4) for visualization, stats (v4.2.1) for
   statistical computations, and car (v3.1–0) for assumption validation;
   statistical significance was determined by One-way ANOVA.

Development and assessment of nomogram

   The multivariate Cox regression model, which includes covariates such
   as clinical stage, initial treatment effect, age, clinical type,
   residual tumor, clinical grade, tumor invasion, radiotherapy, and XPR1
   expression, was used to identify independent prognostic factors in
   clinical pathology. Construct prognostic nomogram, this model was also
   employed to estimate 1-year, 3-year, and 5-year overall survival
   probabilities for EC patients. The model's accuracy was assessed using
   a calibration curve that compared predicted probabilities with actual
   outcomes.Model accuracy was indicated by alignment with reference
   lines, and graphs were produced using the Xiantao Academic Online
   Analysis Tool.The R package used includes survival[3.3.1], rms[6.3–0].

Enrichment analysis of correlated gene sets

   Utilizing TCGA data, we investigated the biological processes and
   pathways associated with XPR1 by analyzing its correlation with other
   EC mRNAs. The significantly co-expressed genes were subjected to GO
   (Gene Ontology, December 2023 version) and KEGG (Release 104.0,
   December 2023 version) pathway analyses by using the Xiantao Academic
   online analysis platform, which incorporates the clusterProfiler
   package as its underlying analytical engine.

Association between XPR1 expression and m6A methylation in EC

   Correlation analyses between XPR1 expression and m6A-related genes
   (YTHDF1, YTHDF2, YTHDF3, YTHDC1, YTHDC2, WTAP, RBM15, RBM15B, ZC3H13,
   HNRNPC, METTL14, METTL3, IGF2BP1, IGF2BP2, IGF2BP3, RBMX, HNRNPA2B1,
   VIRMA, FTO, and ALKBH5) in UCEC were performed by using the ggplot2
   package (v3.4.0) in R software (v4.2.1). Pearson correlation
   coefficients with absolute values > 0.3 and p-values < 0.05 were
   considered statistically significant.

Examination of XPR1 expression and its correlation with immune cell
infiltration in EC

   This study employed a deconvolution method to explore the relationship
   between XPR1 expression and immune cell infiltration in EC by
   estimating immune cell subtype abundance in tumor tissues. Using the
   integrated deconvolution model from the TIMER database alongside TCGA
   expression profiles, we quantified major immune cell subtype
   infiltration (B cells, CD4 + T cells, CD8 + T cells, macrophages,
   neutrophils, and dendritic cells) by resolving mixed gene expression
   signals. Statistical techniques, such as least squares regression, were
   used to determine relative abundance by comparing immune cell-specific
   gene expression patterns with predefined cell subtype-specific
   expression matrices. Standardized parameter settings from the TIMER
   platform ensured analytical reproducibility. Analyses examining
   XPR1-immunological relationships were conducted via single-sample Gene
   Set Enrichment Analysis (ssGSEA) within the Xiantao Academic platform,
   evaluating both: 1) associations between XPR1 expression and immune
   cell subtype infiltration, and 2) differential immune cell expression
   levels between XPR1 high-/low-expression groups (dichotomized at median
   threshold). Results were visualized using ggplot2 (v3.4.4), with
   statistical significance defined as |Pearson r|> 0.3 and P < 0.05. This
   integrated approach delineates XPR1's potential regulatory functions
   within the UCEC tumor immune microenvironment.

Cell culture

   Normal endometrial cells (EEC) were obtained from Hubei Wuhan Pricella
   Co., Ltd., and endometrial cancer cells (ECC-1) from Hunan Fenghui
   Biotechnology Co., Ltd., both in China.All cell lines verified by STR
   analysis prior to experimentation.Cells between passages 5 and 15 were
   used for all experiments to ensure genetic stability. ECC-1 cells were
   cultured in RPMI-1640 medium (Gibco, USA) supplemented with 10% fetal
   bovine serum (Procell, Wuhan, China) and 1% penicillin–streptomycin
   (ABT920; G-clone, Beijing, China). All samples were incubated at 37 °C
   with 5% CO2 in a humidified environment.

western blotting

   Protein extraction and concentration were conducted using the Beyotime
   Biotechnology protein extraction kit (P0033) and BCA kit protocols.
   Following the reagent ratio guidelines, separation and concentration
   gels were prepared, and proteins were loaded for electrophoresis. After
   transferring the proteins to a PVDF membrane, the membrane was blocked
   with 5% skim milk, incubated with a primary antibody (1:1000) at 4 °C
   overnight, and washed with TBST. It was then incubated with a secondary
   antibody at room temperature for 1 h, followed by washing and ECL
   reaction for image capture. GAPDH served as the internal control. The
   primary antibodies used were polyclonal rabbit anti-XPR1 (Abcam,
   ab97483) and rabbit polyclonal to GAPDH (CST, 2118). Band density was
   analyzed using a gel imaging system and compared to the internal
   control, with each experiment performed in triplicate.Image analysis
   was performed using ImageJ.

Lentiviral transfection

   Lentiviral transfection was performed in ECC-1 cells.We purchased XPR1
   silencing lentivirus and overexpression lentivirus (Genepharmacy,
   Shanghai, China). On the first day, T25 culture bottles were seeded
   with target cells at a density of about 60%, at 37 ℃, and 5% CO2
   overnight. The lentivirus dosage was calculated according to the
   Multiplicity of Infection (MOI) = 15.The next day, after diluting the
   lentivirus, add the diluted virus solution to the culture bottle. After
   24–48 h of cultivation, stable cells were screened using puromycin.The
   order of events was the following:
     * Sh-XPR1(Homo,Puro): 5’-GGCCCUUGAUAAGAAUCUATT-3’;
     * NC-Sh-XPR1(Homo,Puro): 5’-TTCTCCGAACGTGTCACGT-3’;

   OE-XPR1(Homo,EF-1a/Puro):The recombinant shuttle plasmid and packaging
   plasmids (pgag/pol, PVV, pVSV-G) were prepared by Genepharmacy
   (Shanghai, China). Plasmid cDNA was extracted using the High Pure
   Plasmid Midiprep Kit (Thermo Scientific™, K0481) and labeled with a
   puromycin (puro)-resistant gene to facilitate subsequent specific
   analysis via puro selection.

EdU proliferation assay

   Cell proliferation was assessed with the Cell-Light TM EdU imaging
   detection kit (Ruibo Biotechnology, Guangzhou, China) according to the
   manufacturer's instructions, and images were analyzed using ImageJ.Each
   group consisted of three replicates, and statistical analysis was
   conducted using a two-tailed t-test, with a p-value below 0.05
   considered statistically significant.

Matrigel invasion assays

   In Transwell invasion assays, a Matrigel basement membrane matrix
   (1 mg/mL, BD Biosciences) was applied to the upper section of an 8 μm
   pore, 6.5 mm polycarbonate Transwell filter membrane (Corning) and
   pretreated for 2 h at 37 °C. Subsequently, 5 × 10^4 ECC-1 cells were
   seeded into the top chamber and incubated for 48 h.Cells that migrated
   to the underside were fixed with 4% paraformaldehyde, stained with 0.5%
   crystal violet (Beyotime), and counted microscopically. Five fields per
   chamber were counted and photographed at 100 × magnification, with each
   experiment performed in triplicate.Image analysis was performed using
   ImageJ.

m6A Dot blot analysis

   RNA from Total EEC and ECC-1 was extracted using TRIzol and quantified
   with a Nanodrop 2000. The sample was heat-denatured at 65 °C for 10 min
   and rapidly cooled on ice. Nylon membranes were treated with 1 µg of
   RNA, UV cross-linked, blocked with 5% non-fat milk, and incubated
   overnight at 4 °C with M6A antibody(Proteintech,68,055–1-Ig). The
   membranes were then incubated with HRP-conjugated goat anti-mouse IgG
   at a 1:2000 dilution and visualized using a chemiluminescence imaging
   system. Membrane loading consistency was assessed by staining with
   0.02% methylene blue for 30 min, followed by rinsing with ultrapure
   water and photographing.Each experiment was conducted in triplicate.

Statistical analysis

   The study analyzed the complete TCGA dataset, including 554 UCEC tumor
   samples and 35 normal samples, to ensure comprehensive and reliable
   statistical outcomes. In vitro experiments involved three biological
   replicates per group, adhering to standard experimental designs for
   detecting intergroup differences. Details on the number of replicates
   are provided in the methods section. As all available samples from
   public databases were used, no pre-statistical sample size estimation
   was performed. Median expression levels were used to categorize XPR1
   gene expression, with scatter plots illustrating its distribution in EC
   patients. Pearson correlation analysis assessed the relationship
   between XPR1 and associated genes. Univariate and multivariate Cox
   regression analyses identified potential prognostic factors.

   Statistical analyses were performed using the Xiantao Academic Online
   Analysis Tool, R (v4.4.2), and GraphPad Prism 9.5, with a two-sided P
   value of less than 0.05 considered significant. Figures were created
   and edited using GraphPad Prism 9.5 and Adobe Illustrator 2022, with
   statistical significance indicated by p-values.Significance levels are
   denoted as follows: * for p < 0.05, ** for p < 0.01, *** for p < 0.001,
   and **** for p < 0.0001.

Results

Expression of XPR1 in EC

   TCGA data analysis showed significant variations in XPR1 expression
   across various cancer types (Fig. [42]1A). In EC samples, both paired
   and unpaired mRNA analyses indicated notably higher XPR1 levels in
   tumors compared to normal tissues (Fig. [43]1B and 1 C). Figure [44]1D
   demonstrates that XPR1 protein levels are significantly elevated in
   cancerous tissues versus adjacent non-cancerous tissues.
   Immunohistochemical data from The Human Protein Atlas corroborated the
   significantly increased XPR1 expression in cancer tissues
   (Fig. [45]1E). Kaplan–Meier survival analysis revealed that higher XPR1
   expression is associated with poorer prognosis (HR = 1.6, 95%
   CI:1.06–2.42, p = 0.025) (Fig. [46]1F). We examined XPR1's genomic and
   copy number variations in EC, analyzing the OncoPrint map using
   cBioPortal and TCGA data (Fig. [47]1G). The findings showed that about
   8% of EC cases had XPR1 gene amplification, missense mutations, or deep
   deletion mutations.

Fig. 1.

   [48]Fig. 1
   [49]Open in a new tab

   Expression of XPR1 in EC. A The expression of XPR1 in Pan cancer was
   analyzed based on TCGA database, (B) and (C) 35 normal samples and 554
   UCEC tumor samples were obtained from TCGA database. After TPM data
   cleaning, the mRNA expression differences between normal samples and
   tumor samples were compared by paired and unpaired methods, (D) the
   protein expression of XPR1 in paracancerous tissues and tumor tissues
   was detected by Western blotting, (E) the immunohistochemical results
   of EC paracancerous tissues and cancer tissues were obtained from he
   human protein atlas database to analyze the expression differences
   between paracancerous tissues and cancer tissues, (F) the effect of
   XPR1 in EC was analyzed based on TCGA database, (G) the cbioportal was
   used.The database was used to analyze the mutation of XPR1 in
   EC.**indicate p < 0.01

The correlation between XPR1 expression levels and clinicopathological
characteristics in EC patients

   We analyzed the TCGA database to investigate the association between
   XPR1 expression and eight clinicopathological features in EC patients,
   including age, BMI, clinical stage, and tumor grade.(G1 low, G2:
   medium, G3: high), disease progression (PD: progressive, SD: stable,
   PR: partial response, CR: complete response), resection status (R0:
   curative, R1: microscopic residual, R2:macroscopic residual
   tumor)(Fig. [50]2A-2L).Our research indicates that increased XPR1
   expression is significantly correlated with age, BMI, clinical stage,
   histological grade, and tumor invasion percentage, particularly in EC,
   and is linked to various clinical characteristics. High XPR1 expression
   adversely affects EC patient prognosis under different
   clinicopathological conditions. Kaplan–Meier analysis using the TCGA
   database revealed that elevated XPR1 expression is associated with
   poorer outcomes, including reduced overall survival, disease-free
   survival, and progression-free interval (Fig. [51]3A-3L, S1.A-F).
   Specifically, in clinical stage I patients, a notable correlation was
   found with overall survival (OS, HR = 3.25, 95% CI: 1.52–6.96,
   p = 0.002), disease-free survival (DSS, HR = 16.04, 95% CI:
   2.11–122.17, p = 0.007), and progression-free interval (PFI, HR = 2.39,
   95% CI: 1.34 − 4.25, p = 0.003). In patients with a BMI ≤ 30,
   significant associations were identified for overall survival
   (HR = 2.58, 95% CI:1.34–4.25, p = 0.005), diabetes (HR = 2.90, 95%
   CI:1.20–6.99, p = 0.018), endometrioid histology (HR = 2.06, 95%
   CI:1.16–3.65, p = 0.014), absence of hormone therapy (HR = 2.10, 95%
   CI:1.19–3.71, p = 0.011), R0 residual tumor status (HR = 2.66, 95%
   CI:1.49–4.74, p = 0.0009), and tumor invasion < 50% (HR = 3.90, 95%
   CI:1.56–9.74, p = 0.004). Conversely, in clinical stage II-IV patients,
   no significant associations were found for overall survival (HR = 1.05,
   95% CI:0.63–1.77, p = 0.842), disease-free survival (HR = 1.00, 95%
   CI:0.57–1.77, p = 0.989), or progression-free interval (HR = 1.12, 95%
   CI:0.71–1.76, p = 0.639). Additionally, no correlation was observed in
   patients with BMI > 30 (HR = 1.15, 95% CI: 0.65 − 2.01, p = 0.634),
   without diabetes mellitus (HR = 1.51, 95% CI: 0.89 − 2.57, p = 0.129),
   mixed and serous histology (HR = 1.39, 95% CI: 0.75 − 2.60, p = 0.298),
   hormone therapy (HR = 3.97, 95% CI: 0.79 − 19.91, p = 0.093), R1 and R2
   residual tumor status (HR = 2.17, 95% CI: 0.78 − 6.02, p = 0.137), and
   tumor invasion ≥ 50% (HR = 1.09, 95% CI: 0.64 − 1.88, p = 0.746).

Fig.2.

   [52]Fig.2
   [53]Open in a new tab

   The correlation between XPR1 expression levels and clinicopathological
   characteristics in EC patients. (A) age, (B) BMI, (C) Clinical stage,
   (D) Histologic grade, (E) Menopause status, (F) Primary therapy
   outcome, (G)Residual tumor, (H) Tumor
   invasion(%),(I)Diabetes,(J)Hormones therapy,(K)OS event,(L)PFI event*,
   ** and *** indicate p < 0.05, p < 0.01 and p < 0.001, respectively

Fig. 3.

   [54]Fig. 3
   [55]Open in a new tab

   Prognostic analysis of XPR1 expression levels and clinicopathological
   features. (A-L)Prognostic subgroup analyses of EC patients with
   different clinicopathological status

Elevated XPR1 expression does not independently affect overall survival risk.

   Univariate and multivariate Cox regression analyses were conducted to
   identify prognostic factors for EC, including variables like clinical
   stage, initial treatment outcome, age, clinical type, residual tumor,
   clinical grading, tumor infiltration, radiation therapy, and XPR1. The
   results indicated that elevated XPR1 expression was not an independent
   predictor of overall survival in EC patients (HR = 1.516, 95%
   CI:0.810–2.839, p = 0.193).Clinical stage (HR = 3.845, 95%
   CI:1.737–8.512, p < 0.001), initial treatment outcome (HR = 0.265, 95%
   CI:0.083–0.848, p = 0.025), and radiation therapy (HR = 0.423, 95%
   CI:0.218–0.820, p = 0.011) were identified as independent prognostic
   factors (Table [56]1). A model incorporating these factors was
   developed and evaluated using calibration curves. The analysis
   demonstrated that clinical stage, primary therapy outcome, and
   radiation therapy correlated with overall survival probabilities at 1,
   3, and 5 years (Fig. [57]4A). Calibration curves, shown in Fig. [58]4B,
   confirmed the model's accuracy by illustrating the alignment between
   observed and predicted values. In conclusion, the model provides a more
   accurate prediction of survival rates for uterine corpus EC) patients.

Table 1.

   Univariate and multivariate Cox regression analyses of clinical
   characteristics associated with OS of UCEC in TCGA
   Characteristcs Total(N) Univariate analysis Multivariate analysis
   Hazard ratio (95% CI) P value Hazard ratio (95% CI) P value
   Clinical stage 553
   Stage I&Stage II 394 Reference Reference
   Stage III&Stage IV 159 3.553 (2.362—5.344)  < 0.001 3.845 (1.737—8.512)
    < 0.001
   Primary therapy outcome 482
   PD 20 Reference Reference
   SD&PR 18 0.705 (0.300—1.658) 0.424 1.333 (0.358—4.958) 0.668
   CR 444 0.110 (0.060—0.204)  < 0.001 0.265 (0.083—0.848) 0.025
   Age 551
    < = 60 207 Reference Reference
    > 60 344 1.850 (1.162—2.944) 0.009 1.636 (0.750—3.571) 0.216
   Histological type 553
   Endometrioid 411 Reference Reference
   Mixed 24 2.428 (1.040—5.672) 0.040 3.097 (0.819—11.704) 0.096
   Serous 118 2.674 (1.744—4.100)  < 0.001 1.245 (0.552—2.811) 0.598
   Residual tumor 414
   R0 376 Reference Reference
   R1 22 1.583 (0.632—3.968) 0.327 2.577 (0.858—7.738) 0.092
   R2 16 5.547 (2.889—10.649)  < 0.001 1.390 (0.479—4.038) 0.545
   Histologic grade 542
   G1&G2 220 Reference Reference
   G3 322 3.298 (1.917—5.672)  < 0.001 1.489 (0.666—3.330) 0.332
   Tumor invasion(%) 475
    < 50 261 Reference Reference
    >= 50 214 2.825 (1.752—4.554)  < 0.001 1.326 (0.629—2.794) 0.458
   Radiation therapy 529
   No 281 Reference Reference
   Yes 248 0.596 (0.387—0.919) 0.019 0.423 (0.218—0.820) 0.011
   XPR1 553
   Low 276 Reference Reference
   High 277 1.602 (1.061—2.418) 0.025 1.516 (0.810—2.839) 0.193
   [59]Open in a new tab

Fig. 4.

   [60]Fig. 4
   [61]Open in a new tab

   High XPR1 expression is not an independent risk factor for overall
   survival. A Nomogram for predicting OS at 1, 3 and 5 years, (B)
   Calibration curves for 1, 3 and 5 years

Functional enrichment of XPR1 related genes

   Using the Limma package, we analyzed differential expression of XPR1 in
   EC, identifying co-expressed genes in the TCGA dataset with |logFC|> 1
   and p.adj < 0.05 (Fig. [62]5A). We identified 7,651 positively and 30
   negatively correlated genes (Pearson correlation > 0.3 and p < 0.05),
   with the top 30 positively correlated genes shown in Fig. [63]5B. The
   ClusterProfiler package facilitated GO term and KEGG pathway enrichment
   analysis of 1,831 genes co-expressed with XPR1. To reduce false
   positives from multiple comparisons, we assessed enrichment analysis
   results using Q values adjusted by FDR correction, defining statistical
   significance as p values < 0.05 and Q values < 0.05. The relevant genes
   were linked to 1,000 biological processes, 260 cell components, 242
   molecular functions, and 57 KEGG pathways. The top five enrichment
   categories in each field were illustrated using bubble charts, with
   significance assessed based on corrected Q values. GO term annotations
   indicate involvement in mRNA processing, histone modification, RNA
   splicing, and regulation of DNA and mRNA metabolism, highlighting
   activity in chromosomal regions, nuclear specks, spindles, spliceosomal
   complexes, and various catalytic functions (Fig. [64]5C–5E). KEGG
   pathway analysis shows associations with Herpes simplex virus 1
   infection, nucleocytoplasmic transport, ubiquitin-mediated proteolysis,
   endoplasmic reticulum protein processing, and RNA degradation
   (Fig. [65]5F).

Fig.5.

   [66]Fig.5
   [67]Open in a new tab

   Functional enrichment of XPR1 related genes. A. Volcano map of
   differential genes, (B) Heat map showing the top 30 genes positively
   associated with XPR1 in EC, (C) Enrichment analyses of BP of
   XPR1co-expressed genes, (D) Enrichment analyses of CC of XPR1
   co-expressed genes, (E) Enrichment analyses of MF of XPR1 co-expressed
   genes, (F) Enrichment analyses of XPR1co-expressed gene terms in KEGG.
   *** indicates p < 0.001

The correlation between XPR1 expression and m6A methylation in EC

   Genes associated with XPR1 are involved in RNA processing, splicing,
   mRNA regulation, and spliceosome formation.The essential function of
   m6A RNA methylation in various RNA processes, such as splicing, export,
   processing, translation, and decay, indicates a possible connection
   with XPR1 expression. In EC, we examined the relationship between XPR1
   expression and 20 m6A-related genes.Within the TCGA-UCEC cohort,
   patients were stratified into XPR1 low-expression (Low) and
   high-expression (High) groups based on a median cutoff of XPR1
   expression values [log₂(TPM + 1)]. This stratification enabled analysis
   of expression correlations between XPR1 and key m6A regulatory genes
   (Pearson r > 0.3, p < 0.001; Fig. [68]6A), including: YTHDF1-3,
   YTHDC1-2, WTAP, RBM15, RBM15B, ZC3H13, HNRNPC, METTL14, METTL3, RBMX,
   HNRNPA2B1, VIRMA, FTO, and ALKBH5. Dot blot analysis showed increased
   m6A methylation levels in ECC-1 cells with high XPR1 expression
   compared to EEC cells (Fig. [69]6B). Scatter plots highlight
   significant correlations (p < 0.001) between XPR1 and 17 m6A-related
   genes, with notable associations for YTHDF1 (R = 0.354), YTHDF2
   (R = 0.589), YTHDF3 (R = 0.614), YTHDC1 (R = 0.621), YTHDC2
   (R = 0.560), WTAP (R = 0.553), RBM15 (R = 0.472), RBM15B (R = 0.401),
   ZC3H13 (R = 0.565), HNRNPC (R = 0.535), METTL14 (R = 0.624), METTL3
   (R = 0.427), RBMX (R = 0.585), HNRNPA2B1 (R = 0.528), VIRMA
   (R = 0.571), FTO (R = 0.459), and ALKBH5 (R = 0.444) (Figs. [70]6C-6M
   and S2.A-F). High levels of YTHDF3 and VIRMA correlate with poor
   prognosis in EC (Figures [71]S2G and S2H), suggesting a regulatory
   relationship among YTHDF3, VIRMA, and XPR1 influencing patient
   outcomes.

Fig. 6.

   [72]Fig. 6
   [73]Open in a new tab

   The correlation between XPR expression level and m6A modification in
   EC. (A)TCGA UCEC cohort analyzed the association between XPR1 and the
   expression of 20 m6A-related genes,(B)Dot bolts. (C-M)Scatter plots
   were plotted to show the association between XPR1 and the expression of
   m6A-related genes including YTHDF1, YTHDF2, YTHDF3, YTHDC1, YTHDC2,
   WTAP, RBM15, RBM15B, ZC3H13, HNRNPC, METTL14

The association between XPR1 expression and immune cell infiltration

   Comprehensive analysis of XPR1-immunocyte infiltration relationships
   revealed distinct correlations between XPR1 expression and specific
   immune subsets in EC.Analyses using the TIMER database revealed that
   XPR1 expression correlated negatively with B cells (partial
   cor = −0.159, p = 0.007) and CD4 + T cells (partial cor = −0.201,
   p = 0.001), while demonstrating positive correlations with CD8 + T
   cells (partial cor = 0.162, p = 0.006) and neutrophils (partial
   cor = 0.136, p = 0.020)(Fig. [74]7A).Similarly, the TCGA database
   analysis showed that XPR1 expression in EC patients was positively
   correlated with central memory T (Tcm) cells (R = 0.465, p < 0.001) and
   helper T (Th) cells (R = 0.315, p < 0.001), but negatively correlated
   with NK CD56dim cells (R = − 0.349, p < 0.001), cytotoxic cells (R =
   − 0.355, p < 0.001), NK CD56bright cells (R = − 0.366, p < 0.001), and
   plasmacytoid dendritic cells (pDC) (R = − 0.426,
   p < 0.001)(Fig. [75]7B).A heatmap depicted interactions among
   tumor-infiltrating immune cells (TIICs) in tumor samples(Fig. [76]7C).
   Comparative assessment of immune cell subpopulations between XPR1-high
   and XPR1-low expression groups demonstrated XPR1-associated alterations
   in the distribution of multiple immunocyte subsets within the EC
   microenvironment(Fgure7D,S3G-L). These alterations encompassed:
   activated dendritic cells (aDC), B cells, CD8⁺ T cells, cytotoxic
   lymphocytes, conventional dendritic cells (DC), immature dendritic
   cells (iDC), neutrophils, NK CD56bright cells, NK CD56dim cells,
   plasmacytoid dendritic cells (pDC), T cells, T helper cells, central
   memory T cells (Tcm), effector memory T cells (Tem), follicular helper
   T cells (TFH), gamma delta T cells (Tgd), T helper 1 cells (Th1), T
   helper 17 cells (Th17), T helper 2 cells (Th2), and regulatory T cells
   (Treg). Collectively, these observations indicate XPR1's potential role
   in modulating immune microenvironment dynamics in EC.

Fig. 7.

   [77]Fig. 7
   [78]Open in a new tab

   The correlation between XPR1 expression and immune infiltration. (A)The
   timer database shows the correlation between XPR1 and related immune
   cells in EC,(B) The correlation between XPR1 and immune infiltrating
   cells in EC,(C) heatmap shows the relationships among
   tumor-infiltrating immune cells (TIICs),(D) Differential distribution
   of immune cells in patients with high XPR1 expression and low XPR1
   expression.*, ** and *** indicate p < 0.05, p < 0.01 and p < 0.001,
   respectively

XPR1 facilitates the proliferation and dissemination of UCEC cells

   This study examined how altering XPR1 expression affects ECC-1 cell
   proliferation and invasion in EC.ECC-1 cells were divided into four
   groups: knockdown control (NC-Sh-XPR1,Sh-Scramble), knockdown
   (Sh-XPR1), overexpression control (NC-OE-XPR1,OE-Scramble), and
   overexpression (OE-XPR1).Genetic manipulation via XPR1 knockdown
   (Sh-XPR1) and overexpression (OE-XPR1) confirmed dose-dependent XPR1
   modulation relative to respective controls (NC-Sh-XPR1 and NC-OE-XPR1,
   Fig. [79]8A-8B).Cell proliferation was assessed using the EdU
   proliferation kit.The study found that XPR1 knockdown significantly
   reduced ECC-1 cell proliferation (p = 0.002, −0.07112 ± 0.009951),
   while XPR1 overexpression significantly increased proliferation
   (p < 0.0001, 0.1819 ± 0.01077) (Fig. [80]8C-8D).The Transwell invasion
   assay demonstrated that XPR1 knockdown significantly decreased cell
   invasiveness (p = 0.0004, −43.33 ± 3.844), whereas its overexpression
   significantly enhanced invasiveness (p = 0.0017, 61.67 ± 8.192)
   (Fig. [81]8E-8F). These findings suggest that XPR1 plays a crucial role
   in the proliferation and invasion of EC cells, consistent with
   correlation analysis results, highlighting its importance in EC
   progression.

Fig. 8.

   [82]Fig. 8
   [83]Open in a new tab

   XPR1 enhances the proliferation and metastasis of EC cells. A and B
   Validation of XPR1 protein expression knockdown and overexpression in
   ECC-1cells. C and D after silencing and overexpressing XPR1 in ECC-1
   cells, the proliferation of ECC-1 cells were detected by EdU, (E) and
   (F) after silencing and overexpressing XPR1 in ECC-1 cells, the
   invasion of ECC-1 cells cells were detected by Transwell.**, ***and****
   indicate p < 0.01, p < 0.001,andp < 0.0001 respectively

Discussion

   EC is the most common cancer in the female reproductive system and is
   on the rise globally [[84]15]. Although treatments like surgery,
   chemotherapy, and radiation therapy have advanced, many patients still
   face poor prognoses due to difficulties in early diagnosis and precise
   prognostic evaluation [[85]16]. This highlights the urgent need for
   novel biomarkers to improve treatment outcomes and strategies for
   patients.

   XPR1 is significantly overexpressed in various cancers and is
   associated with poor patient prognosis [[86]12]. It facilitates the
   efflux of intracellular inorganic phosphorus (PI), maintaining cellular
   phosphate balance [[87]15]. This study systematically reviewed and
   integrated the research design, data analysis methods, and experimental
   validation to ensure alignment with the overall research framework.
   Utilizing the TCGA database as the primary data source, the study also
   incorporated data from cBioPortal, the Human Protein Atlas, and other
   public databases to comprehensively analyze XPR1 expression in EC and
   its clinical correlations.The study utilized various analytical
   methods, including survival analysis (Kaplan–Meier and Cox regression),
   prognostic nomogram construction, gene function enrichment analysis (GO
   and KEGG), and correlation analyses of immune infiltration and m6A
   modification-related genes, to ensure comprehensive findings. Database
   analysis results were validated through Western blot and cell function
   tests (EdU proliferation and Transwell invasion tests), enhancing the
   research's reliability. TCGA data analysis indicated high XPR1
   expression in EC tissues, correlating with clinicopathological features
   such as age, BMI, stage, grade, and tumor invasion. Cell experiments
   showed that XPR1 expression changes significantly affected EC cell
   proliferation and metastasis. Kaplan–Meier and Cox regression analyses
   confirmed the link between high XPR1 expression and poor prognosis in
   EC patients. Gene function enrichment analysis identified that genes
   coexpressed with XPR1 are mainly involved in RNA processing, DNA
   metabolism, and other critical biological processes, providing a
   theoretical basis for understanding XPR1's role in tumor development.
   The analysis of m6A RNA methylation modification and immune
   infiltration highlights XPR1's potential role in regulating the tumor
   microenvironment, suggesting its involvement in EC progression through
   m6A modification and immune regulation.

   Studies reveal that XPR1 is overexpressed in esophageal cancer
   [[88]17]. Gene expression analysis shows that XPR1 mRNA level in ucec
   tissues is significantly higher than that in normal endometrial tissues
   (p < 0.001).XPR1's expression across various cancer types suggests its
   potential as a biomarker for early diagnosis and targeted therapies in
   EC, emphasizing its role in tumorigenesis beyond this cancer. Combining
   XPR1 with other biomarkers may enhance diagnostic precision and patient
   outcomes.

   Kaplan–Meier analysis revealed that elevated XPR1 expression correlated
   with poorer overall survival in UCEC patients (HR = 1.60, 95% CI:
   1.06–2.42, p = 0.025). However, multivariate Cox regression analysis
   demonstrated that increased XPR1 expression did not constitute an
   independent prognostic factor for overall survival (HR = 1.52, 95% CI:
   0.81–2.84, p = 0.193).These findings suggest that XPR1 may not
   independently influence patient survival through direct mechanisms, but
   rather contributes to prognosis through its correlations with other
   clinicopathological covariates and underlying biological
   characteristics.We identified significant correlations between XPR1
   expression levels and multiple adverse clinical parameters, including
   patient age, BMI, clinical stage, histological grade, and tumor
   invasion proportion. Concurrent associations were observed with immune
   cell infiltration (including but not limited to B cells, CD8⁺ T cells,
   and CD4⁺ T cells) and N6-methyladenosine (m6A) methylation
   modifications.These findings suggest that XPR1 may indirectly influence
   EC patient survival outcomes through potential modulation of tumor
   microenvironment dynamics, facilitation of tumor cell proliferation and
   invasion, and/or regulation of immune responses and epigenetic
   modifications.For example,Elevated XPR1 expression frequently coincided
   with adverse clinicopathological manifestations—including advanced
   disease stage, deteriorating histological grade, and increased tumor
   invasion depth—each constituting established prognostic contributors
   strongly associated with unfavorable outcomes in EC.

   XPR1 expression correlates with immune cell infiltration, particularly
   CD8 + T cells and B cells, highlighting its role in the EC immune
   microenvironment and its potential impact on tumor immunity and
   immunotherapy response. Investigating XPR1's interaction with immune
   cells may lead to novel therapeutic approaches to boost anti-tumor
   immunity. Pathway analysis indicates that genes co-expressed with XPR1
   are involved in key biological processes and signaling pathways, such
   as RNA processing and DNA metabolism, suggesting a connection between
   XPR1 and tumor-related pathways and clinical outcomes. Understanding
   how XPR1 influences these pathways is essential for further insights.

   The link between XPR1 and m6A methylation regulators indicates that
   XPR1 might be involved with post-transcriptional gene expression
   regulation in EC. XPR1 shows a significant correlation with m6A-related
   genes, suggesting that XPR1 may be related to the process of m6A
   modification but its direct effect on the stability and translation of
   m6A-modified mRNAs remains to be fully elucidated and needs further
   investigation. Such an impact could influence tumor cell behavior and
   treatment response.

   This study has notable limitations. Public databases like The Cancer
   Genome Atlas (TCGA) may experience batch effects during sample
   collection, processing, and sequencing. Although efforts were made to
   minimize these effects, conclusions require further validation with
   larger, independent cohorts and experimental studies. Additionally, the
   study does not establish XPR1's role in the causal mechanisms of tumor
   biology. The limited sample size of 554 tumor samples and 35 normal
   samples may affect the generalizability of the findings, highlighting
   the need for validation in diverse patient cohorts. Future research
   should prioritize experimental validation and larger, independent
   cohorts to ascertain the clinical relevance of XPR1 in uterine corpus
   EC.

   This study confirmed that high XPR1 expression is associated with poor
   prognosis in EC patients, but failed to demonstrate its role as an
   independent prognostic factor. These findings suggest that the
   prognostic significance of XPR1 may partially stem from its correlation
   with other conventional clinical characteristics (such as stage, grade,
   etc.). The connection between XPR1 and immune cell infiltration
   emphasizes its role in the tumor microenvironment, presenting
   opportunities for novel immunotherapeutic approaches. Despite certain
   limitations, the findings strongly advocate for XPR1's involvement in
   EC pathophysiology and promote further investigation into its molecular
   mechanisms and therapeutic potential. Targeting XPR1 could improve
   diagnostic and treatment strategies, potentially enhancing clinical
   outcomes for EC patients.

Supplementary Information

   [89]Supplementary Material 1^ (398.8KB, docx)
   [90]Supplementary Material 2^ (1.5MB, docx)

Acknowledgements