PLA2G2A⁺ CAFs as a Key Driver of Lymph Node Metastasis in Penile Cancer and Its Underlying Mechanisms

PLA2G2A⁺ CAFs as a Key Driver of Lymph Node Metastasis in Penile Cancer and Its Underlying Mechanisms | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article PLA2G2A⁺ CAFs as a Key Driver of Lymph Node Metastasis in Penile Cancer and Its Underlying Mechanisms hu chen, chen weijia, Li fan, feng minghong, wang jianglai, tan zijian, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8936508/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Metastasis to lymph nodes in penile squamous cell carcinoma (PSCC) significantly impacts patient survival rates, and the challenge of accurately identifying patients who would genuinely benefit from inguinal lymph node dissection prior to surgery remains a pressing and unresolved issue in clinical practice. This research utilized single-cell RNA sequencing, bulk RNA sequencing, and various machine learning techniques (including LASSO, Random Forest, and XGBoost) to systematically identify PLA2G2A as a crucial molecular marker for lymph node metastasis in PSCC. In a separate clinical cohort of 69 patients, immunohistochemical studies demonstrated that PLA2G2A was markedly elevated in those with metastasis, and its high levels correlated with unfavorable clinical outcomes. Additionally, we pinpointed PLA2G2A expression to a distinct subpopulation of fibroblasts. Mechanistic investigations indicated that PLA2G2A-positive cancer-associated fibroblasts (CAFs) attract CXCR4-positive monocytes into the tumor microenvironment via CXCL12 secretion, subsequently driving their transformation from an M1-like to an M2-like immunosuppressive phenotype through the MDK–NCL/LRP1 signaling pathway, thus fostering a tumor microenvironment conducive to metastasis. Overall, this study clarifies the significant regulatory function of PLA2G2A-positive CAFs in PSCC metastasis and positions PLA2G2A as a valuable biomarker for evaluating the risk of lymph node metastasis before surgery. Health sciences/Biomarkers Biological sciences/Cancer Health sciences/Oncology Penile Cancer PLA2G2A Lymph Node Metastasis Cancer-Associated Fibroblasts Biomarker Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 INTRODUCTION Penile cancer is an uncommon type of cancer, exhibiting significant differences in occurrence among various global populations. In some areas of Africa, Asia, and South America, it can represent as much as 10% of all cancers affecting men (1, 2) . In China, the current rate of penile cancer is estimated to be around 0.5 to 2.5 cases per 100,000 individuals (3) . While the overall frequency is low, advanced stages of this disease can be life-threatening and greatly diminish the quality of life for those affected. Treatments for penile cancer frequently lead to considerable anatomical and functional impairments in the genital region, resulting in issues such as urinary and sexual dysfunction, along with considerable psychological distress due to changes in appearance. (4) . Consequently, even though it is a rare form of cancer, penile cancer deserves increased public awareness, with a focus on improving treatment options and enhancing the long-term well-being of patients. Penile cancer is an uncommon type of cancer, especially infrequent in advanced Western nations. This rarity has led to insufficient extensive and high-caliber foundational studies, along with considerable obstacles in executing randomized phase III clinical trials. As a result, there are notable deficiencies in both translational and basic research within this area. Yunnan Province in China, known for its economic challenges, experiences a notably high rate of penile cancer. This situation is linked to various risk factors, such as a significant number of smokers, inadequate hygiene that causes chronic inflammation due to smegma buildup, and a widespread occurrence of HIV, which correlates with higher HPV infection rates. Together, these elements raise the likelihood of penile cancer in the area. Consequently, the increased prevalence of this disease in Yunnan creates advantageous and concrete opportunities for conducting penile cancer research at our facility. The prognosis for penile cancer shows significant variation. For men with localized penile cancer (N0 stage), the 5-year survival rate is around 90% (5) . However, the situation worsens with the onset of lymph node metastasis. Patients with unilateral inguinal lymph node involvement of two or fewer nodes (N1 stage) have a 5-year survival rate of about 80%, but this figure plummets to between 10% and 20% for those with bilateral inguinal or pelvic lymph node involvement (N2 or N3 stage). In instances where there is extranodal lymph node spread, the overall survival rate drops below 10% (6) . Consequently, the presence of multiple lymph node metastases (indicative of advanced penile cancer) leads to a significant decline in patient survival rates. Timely identification and intervention are essential factors influencing patient survival rates, with lymph node metastasis being significantly linked to unfavorable outcomes. Research indicates that performing early inguinal lymph node dissection can improve survival chances (7) . Nonetheless, this procedure often results in postoperative issues, particularly lymphatic leakage, which may require extended drainage periods. Furthermore, patients may experience complications like lymphedema in the lower limbs, infections or necrosis of the surgical site, and hernia development, all of which can adversely affect their long-term quality of life (8) . Currently, guidelines from the NCCN, EAU, and Chinese authorities advocate for immediate bilateral inguinal lymph node dissection when enlarged inguinal lymph nodes are detected, irrespective of whether penile cancer is categorized as low, intermediate, or high risk. Notably, around 30% of intermediate-risk patients without clinically evident lymphadenopathy are discovered to have hidden micrometastases upon postoperative pathological assessment (9) . Conversely, some patients exhibiting palpable inguinal lymphadenopathy do not show any metastatic involvement in their postoperative pathology results. This indicates a lack of sufficiently sensitive biomarkers or imaging techniques to reliably predict the existence of hidden micrometastases in the inguinal lymph nodes. Consequently, certain patients may experience overtreatment, leading to avoidable complications, while others may not receive adequate treatment, resulting in disease advancement. Thus, determining the right candidates for inguinal lymph node dissection in early-stage penile cancer is a pressing and unresolved clinical issue. This study seeks to combine single-cell transcriptomic analysis with machine learning methods to pinpoint molecular markers that could aid in the preoperative assessment of lymph node metastasis. These markers could enhance clinical decision-making regarding the necessity of inguinal lymph node dissection, helping to prevent overtreatment and its complications, as well as avoiding disease progression due to delayed or missed surgical interventions. MATERIALS AND METHODS Single-cell RNA-seq Data Acquisition and Construction of Atlas The initial raw data for single-cell RNA sequencing (scRNA-seq) was supplied by Maarten Albersen's group at University Hospitals Leuven (10) , aligning with earlier reports. Transcriptomic data along with clinical details for patients with esophageal cancer were retrieved from The Cancer Genome Atlas (TCGA). For external validation, additional cohorts were acquired from the Gene Expression Omnibus (GEO). The development of the Single-cell Atlas involved data analysis through the Seurat R package (version 4.0). Initial Seurat objects were created by applying filtering criteria that removed genes found in fewer than three cells and cells with fewer than 200 expressed features. After merging the datasets, quality control was conducted by evaluating the proportions of mitochondrial (pMT), hemoglobin (pHB), and ribosomal (pRP) genes. The data underwent normalization using the NormalizeData function, and the top 2,000 most variable features were selected. A linear transformation was executed with ScaleData (Z-score scaling), and batch effects were adjusted using the Harmony algorithm (RunHarmony). Dimensionality reduction was achieved through PCA, selecting the top 30 principal components for UMAP embedding. Cell clusters were determined using FindNeighbors and FindClusters (resolution set to 0.5). Differential expression analysis was performed with FindAllMarkers (logFC > 0.25, only.pos = TRUE), and cell types were annotated according to established markers from existing literature and public databases. Sample Population and Ethics Statement This forward-looking investigation involved individuals diagnosed with PSCC at Yunnan Cancer Hospital in Kunming, China, from September 2022 to April 2025. Out of 137 participants who satisfied the criteria for inclusion and exclusion, 77 had comprehensive follow-up information, while 69 had accessible formalin-fixed paraffin-embedded (FFPE) tissue samples. Informed written consent was obtained from these 69 patients, and the Medical Ethics Committee of Yunnan Cancer Hospital approved the study protocol (KYLX2022051) prior to the commencement of data collection. The research adhered to the principles outlined in the Declaration of Helsinki (2013 revision). Inclusion criteria are as follows: 1) Diagnosed with PSCC by pathology; 2) Willing to participate in the research protocol and willing to follow up. The exclusion criteria are as follows: 1) non-squamous cell carcinoma or the presence of other histological differentiation; 2) no lymph node dissection performed; 3) Lack of complete FFPE tissue blocks; 4) Patients with incomplete medical records or unable to obtain required clinical information; 5) Patients unwilling to provide clinical information. Immunohistochemistry (IHC) and Scoring Analysis Immunohistochemical analysis was conducted on tissue samples from 44 patients without lymph node involvement and 25 patients with lymph node metastasis. The IHC analysis utilized an antibody targeting PLA2G2A (1:200; Cat# PA5-102403, Thermo Fisher Scientific). Tissue slides were sourced from Yunnan Cancer Hospital, with collection occurring between November 2023 and August 2024. Informed consent was obtained from all participants, and the research protocol received approval from the Medical Ethics Committee at Yunnan Cancer Hospital prior to the commencement of data gathering. Microscopic images of sections were captured from five randomly selected fields. The protein levels were evaluated using the IHC profiler plugin in ImageJ. The scoring for the percentage of positively stained cells was based on these criteria: 0 for less than 10% positive cells, 1 for 10–25%, 2 for 26–50%, 3 for 51–75%, and 4 for more than 75%. Staining intensity was rated as follows: 0 for no staining, 1 for weak, 2 for moderate, and 3 for strong staining. The overall staining score was determined by multiplying the intensity score by the proportion score, yielding a total that could range from 0 to 12. Survival analysis The analysis employing Kaplan-Meier was conducted with the R packages ‘survival’ (version 3.1–8) and visualized through ‘survminer’ (version 0.4.9). A detection threshold of 95% specificity was applied, with the exception of overall and recurrence-free survival calculations, which utilized a threshold of 99% specificity. The survival curves were derived from the overall survival data of patients, adhering to the 99% specificity threshold. Statistical Analysis Data collection and analysis were performed utilizing SPSS version 26.0 (IBM Corp., Armonk, NY, USA) and R version 4.2.1 (R Core Team, R Foundation for Statistical Computing, Vienna, Austria). The lymph node count is presented as mean ± standard deviation, and group comparisons were conducted using the t-test. A p-value of less than 0.05 was deemed statistically significant. RESULTS Construction of the Single-Cell Transcriptomic Atlas for Penile Cancer In order to thoroughly explore the diversity within the tumor microenvironment (TME) of penile cancer (PeCa) and identify potential changes linked to lymph node metastasis (LNM), we examined a single-cell RNA sequencing dataset provided by Maarten Albersen's team at University Hospitals Leuven. The study included 17 patients, divided into two groups: 8 without lymph node metastasis (LNM-) and 9 with lymph node metastasis (LNM+). After applying stringent quality control measures, we selected 60,631 high-quality single cells for further analysis. The application of Uniform Manifold Approximation and Projection (UMAP) for dimensionality reduction resulted in the classification of these cells into 27 unique clusters ( Fig. 1 A ) . Utilizing the expression profiles of key marker genes ( Fig. 1 E ) , the identified clusters were classified into ten primary cell types ( Fig. 1 B ) . These classifications encompassed epithelial cells (characterized by KRT5hi, TP63hi, SFNhi), T lymphocytes (marked by CD3Dhi, CD3Ehi), fibroblasts (identified by COL1A1, DCN, LUM), myeloid cells (indicated by LYZ, CD68), endothelial cells (noted by PECAM1, VWF), natural killer (NK) cells (defined by GNLY, NKG7), B lymphocytes (identified by CD79A, MS4A1), plasma cells (characterized by IGHG1, JCHAIN), mast cells (marked by TPSB2, KIT), and smooth muscle (SM) cells (indicated by ACTA2, TAGLN). To validate the strength of our analysis, we assessed possible batch influences among the samples. The UMAP visualization revealed a consistent cell distribution throughout all clusters, regardless of the patient's origin ( Fig. 1 C ) or metastatic condition (LNM- vs. LNM+) ( Fig. 1 D ) , suggesting successful data integration with minimal bias related to batch effects. In conclusion, we examined the cellular makeup of the PeCa environment. Epithelial cells, T cells, and fibroblasts emerged as the main cell types present in the tumor microenvironment (TME) ( Fig. 1 F ) . Importantly, a comparison between the LNM- and LNM+ cohorts showed no significant differences in the relative proportions of the key cell types (Supplementary Fig. 1) . This indicates that the process of lymph node metastasis in penile cancer may not be primarily influenced by the proliferation of certain cell lineages, but instead by changes in the functional states and inherent molecular traits of particular subpopulations. This insight calls for a more thorough exploration of the molecular features at the level of specific cell subsets. Identification of PLA2G2A as a Predictive Marker for Lymph Node Metastasis by Machine Learning In order to identify strong gene features linked to lymph node metastasis (LNM), we utilized three distinct machine learning techniques: LASSO, Random Forest, and XGBoost. The analysis of Receiver Operating Characteristic (ROC) curves revealed excellent predictive capabilities for each model, with Area Under the Curve (AUC) scores of 0.960, 0.931, and 0.962, respectively ( Fig. 2 A ) . Additionally, confusion matrices validated the high classification precision of these approaches ( Fig. 2 B ) . Following this, we identified the 20 most significant genes from each method ( Fig. 2 C ) . To determine the most dependable biomarker, we examined the overlap among these three gene collections. The Venn diagram analysis indicated that PLA2G2A was the only gene consistently found in the highest-ranked lists across all three methods ( Fig. 2 D ) . In order to confirm this result, we analyzed gene expression patterns in both metastatic and non-metastatic samples. The volcano plot analysis revealed a notable increase in PLA2G2A levels within the LNM+ group ( Fig. 2 E ) . This finding was additionally supported by violin plots, which demonstrated a marked rise in PLA2G2A expression in metastatic samples when compared to LNM- controls ( Fig. 2 F ) . To ensure independent validation, we conducted bulk RNA sequencing on an additional group of four patients with penile cancer (two with lymph node metastasis and two without) from Yunnan Cancer Hospital. In line with our findings from single-cell analysis, the differential expression assessment revealed a notable increase in PLA2G2A levels in the group with lymph node metastasis ( Fig. 2 G ) . In conclusion, we charted the expression levels of PLA2G2A across our single-cell atlas. The feature plots indicated that PLA2G2A was mainly confined to fibroblast clusters, showing minimal expression in other cell types ( Fig. 2 H ) . This unique distribution underscores the significance of the stromal compartment, leading us to concentrate our further research on the various fibroblast subpopulations. PLA2G2A are predictive markers for lymph node metastasis A cohort of 69 patients diagnosed with PSCC underwent immunohistochemistry (IHC) analysis. Comprehensive baseline information is presented in Table 1 . There were no notable differences in age, smoking habits, body mass index (BMI), tumor morphology, lesion site, size, presence of distant metastases, type of surgical intervention, tumor grade, triglyceride levels, total cholesterol, HDL cholesterol, LDL cholesterol, fibrinogen, or squamous cell carcinoma antigen between the groups with and without lymph node involvement. However, significant variations were noted in pathological T stage, overall survival, and recurrence rates between the two groups. The expression of PLA2G2A was significantly elevated in the lymph node-positive group (p < 0.05) ( Fig. 3 ) . Spearman correlation analysis of different clinical pathological features between the N0 and N1 groups Through the analysis of the relationships among the clinical characteristics mentioned earlier, it was initially determined that the pathological T stage, vital status, and recurrence status were linked to lymph node metastasis. In contrast, the other clinical features did not exhibit any notable correlations. Comprehensive findings can be found in Table 2. Correlation of PLA2G2A With Clinical Outcomes We categorized cases with IHC scores exceeding 0 for PLA2G2A as the positive cohort (P+), while those with an IHC score of 0 were designated as the negative cohort (P−). Among the 69 patients followed for a median duration of 41 months (interquartile range: 12.6–62.8 months), 14 had recurrences and 13 succumbed. The Kaplan-Meier survival analysis indicated that positivity for PLA2G2A did not correlate with overall survival (hazard ratio [HR] = 1.21; 95% CI: 0.4026–3.637, p > 0.05). However, it was linked to poorer disease-free survival (DFS) (hazard ratio [HR] = 3.332; 95% CI: 1.156–9.605, p = 0.0259) ( Fig. 4 ). Identification of a PLA2G2A + CAF Fibroblast Subpopulation Associated with Lymph Node Metastasis To explore the precise function of PLA2G2A in the lymph node metastasis of penile cancer, we conducted a detailed sub-clustering analysis of the fibroblast population. Through unsupervised clustering, we identified 14 unique initial clusters of fibroblasts ( Fig. 5 A ) . These clusters were subsequently classified into six primary functional subtypes based on the expression patterns of key marker genes: PLA2G2A+ cancer-associated fibroblasts (CAFs), inflammatory CAFs (iCAFs), myofibroblastic CAFs (myCAFs), progenitor-like fibroblasts, tissue-specific fibroblasts, and various other types ( Fig. 5 B ) . An examination of the marker genes within each subcluster indicates that the PLA2G2A+ subpopulation is distinguished not only by the notable increase in PLA2G2A expression but also by elevated levels of the chemokine CXCL12 and ADH1B ( Fig. 5 C ) . We then categorized the fibroblast atlas based on metastatic status. UMAP visualization revealed a significant concentration of the PLA2G2A+ subpopulation in the LNM+ cohort (Supplementary Fig. 2) . Additionally, the spatial pattern of PLA2G2A expression closely corresponded with this cluster associated with metastasis (Supplementary Fig. 3) , further emphasizing the strong link between this subpopulation and the metastatic characteristics. To clarify the functional characteristics of the PLA2G2A+ subpopulation, we conducted a Gene Ontology (GO) enrichment analysis on its defining genes. The results indicated a notable enrichment in pathways associated with the organization of the extracellular matrix (ECM), the structuring of extracellular components, and the proliferation of epithelial cells ( Fig. 5 D ) . This subpopulation demonstrated not only strong capabilities for matrix remodeling but also elevated levels of various secretory factors, such as CXCL12. This pronounced secretory function aligns with the identified enrichment in the \response to endoplasmic reticulum stress\ pathway. These results imply that PLA2G2A+ CAFs could facilitate lymph node metastasis by creating a specialized physical environment or by releasing chemokines that attract and reprogram immune cells, including macrophages, within the tumor microenvironment. CAF-derived PLA2G2A facilitates the attraction of monocytes through CXCL12 and promotes the transition from M1 to M2 macrophages via the MDK-NCL/LRP1 pathway. To thoroughly analyze how PLA2G2A+CAFs alter the tumor microenvironment (TME) through intercellular interactions—particularly regarding the myeloid compartment—we conducted an extensive ligand-receptor interaction study utilizing CellChat. We initially identified PLA2G2A+ CAFs as the signaling origin to chart their overall communication dynamics. The circle plot visualization indicated that the most significant interaction strengths were observed between PLA2G2A+ CAFs and various lymphocytes (including T cells, B cells, and NK cells) as well as myeloid cells ( Fig. 6 A ) . At the ligand-receptor interaction level, the CXCL12-CXCR4 pathway was widely recognized between CAFs and multiple immune cell types, indicating that PLA2G2A+ CAFs have a broad chemotactic influence ( Fig. 6 B ) . Notably, we detected a pronounced and specific enrichment of the CSF1-CSF1R pathway in interactions between CAFs and myeloid cells ( Fig. 6 B ) . Since CSF1 is crucial for the survival, proliferation, and differentiation of macrophages, this observation highlights that PLA2G2A+ CAFs not only attract immune cells but also create a specialized ecological niche that fosters the survival and presence of myeloid populations. To clarify the unique effects of signals from cancer-associated fibroblasts (CAFs) on various myeloid differentiation states, we isolated and re-categorized myeloid cells into eight distinct sub-groups ( Fig. 6 C ) . Utilizing established marker profiles ( Fig. 6 D ) , we identified these groups as: Monocytes (FCN1+, VCAN+), M1 Macrophages (CXCL9+, CXCL10+, IL1B+), M2 Macrophages (CD163+, FOLR2+, TREM2+), Transitional Tumor-Associated Macrophages (TAMs) (C1QC+, APOE+, TMSB4X+), Proliferating TAMs (MKI67+, TOP2A+), along with other myeloid elements such as Neutrophils (CSF3R+, FCGR3B+), Langerhans Cells (CD207+, CD1A+), and Dendritic Cells (identified by cDC/pDC markers). This detailed classification established a framework for understanding the specific spatiotemporal dynamics of CAF signaling. By examining the interactions between PLA2G2A+ CAFs and various myeloid subpopulations ( Fig. 6 E ) , we mapped out a complex signaling network that encompasses both recruitment and polarization processes. During the recruitment stage, we found that the chemokine receptor CXCR4, which binds to CXCL12, was significantly present in non-TAM myeloid groups ( Fig. 6 E ) . Our pathway analysis identified PLA2G2A+ CAFs as the primary source of CXCL signals ( Fig. 6 F ) , suggesting that these CAFs mainly rely on CXCL12 to effectively draw precursor monocytes from the bloodstream into the tumor microenvironment, thus kickstarting immune infiltration. Our findings on polarization revealed a crucial mechanism through which cancer-associated fibroblasts (CAFs) influence the phenotypic shift of macrophages. Midkine (MDK), a growth factor that binds heparin and is secreted, is recognized for its role in steering tumor-associated macrophages towards an immunosuppressive M2 phenotype by activating receptors like Nucleolin (NCL) and LRP1. Analysis of receptor expression indicated that M1 macrophages, which are generally linked to anti-tumor activity, exhibited a notable presence of the MDK receptors NCL and LRP1 ( Fig. 6 E ) . Additionally, CellChat analysis highlighted a significant signaling interaction involving MDK-NCL and MDK-LRP1 between PLA2G2A+ CAFs and M1 macrophages, with an intensity similar to that found in M2 macrophages ( Fig. 6 E ) . A comprehensive examination of the MK signaling pathway network ( Fig. 4 G ) and the heatmaps depicting signal roles ( Fig. 6 H ) validated that PLA2G2A+ CAFs are the main source of MDK signals, while both M1 and M2 macrophages act as crucial recipients. The gathered evidence indicates that PLA2G2A+ CAFs perform a synchronized two-phase regulatory role: first, they attract monocytes through CXCL12, and then they engage the NCL/LRP1 receptors on M1 macrophages through the MDK signaling pathway. This ongoing signaling is likely to trigger changes in gene expression that suppress the pro-inflammatory activities of M1 and encourage a shift towards the M2 phenotype, thus creating and maintaining an immunosuppressive environment that supports tumor advancement. Pseudotime Trajectory Analysis Corroborates CAF-Driven Differentiation of Monocytes toward an M2-Like Phenotype via an M1-Like Intermediate To substantiate our hypothesis derived from ligand-receptor interaction analysis—specifically, that PLA2G2A+ CAFs recruit monocytes and orchestrate their transition from an M1-like state toward an M2-like phenotype—we performed pseudotime trajectory analysis on the identified monocyte and macrophage populations using Monocle 3. The inferred trajectory recapitulated a continuous developmental lineage, anchoring Monocytes at the root of the pseudo-temporal axis. As differentiation progressed, the trajectory revealed that monocytes evolved through an intermediate state composed of Transitional TAMs and M1 Macrophages, eventually culminating in terminal differentiation into M2 Macrophages, alongside a minor branch of Proliferating TAMs ( Fig. 7 A ) . This topological structure provides compelling computational evidence supporting a sequential differentiation model wherein recruited monocytes initially acquire M1-like characteristics but are ultimately reprogrammed into M2 macrophages within the tumor microenvironment. To dissect the molecular dynamics underlying this M1-to-M2 transition, we interrogated the dynamic expression profiles of canonical marker genes along the pseudo-time axis ( Fig. 7 B, 7 C ). Aligning with the recruitment model, monocyte markers such as FCN1 and the CXCL12 receptor CXCR4 exhibited maximal expression at the trajectory root followed by a rapid decline, confirming that chemotactic responsiveness is a transient feature of early-stage monocytes. Crucially, analysis of polarization markers unveiled a dynamic "phenotype switching" event: the M1-associated proinflammatory chemokine CXCL9 displayed a distinct, transient surge during the intermediate phase (spanning Transitional TAMs and M1 Macrophages), followed by downregulation. In sharp contrast, M2-associated immunosuppressive markers—including CD163, FOLR2, and TREM2—exhibited progressive and sustained upregulation, peaking at the trajectory terminus. This pattern—characterized by the transient induction and subsequent suppression of M1 features coupled with the progressive accumulation of M2 traits—transcriptionally validates the reprogramming process from a proinflammatory to an immunosuppressive phenotype. Furthermore, to validate the mechanistic drivers of this reprogramming, we mapped the expression of receptors for CAF-secreted ligands onto the trajectory. NCL (a receptor for MDK) maintained expression throughout the differentiation lineage, whereas LRP1 (another MDK receptor) and CSF1R (receptor for CSF1) showed a marked resurgence or elevation specifically at the terminal M2 stage (Fig. 5 B). The synchronized upregulation of these receptors with the acquisition of the M2 phenotype suggests that differentiating macrophages, under the regulatory influence of escalating MDK and CSF1 signals derived from PLA2G2A+ CAFs, undergo an initial differentiation toward an M1 stage before ultimately and stably transitioning into the M2 phenotype along the developmental trajectory. DISCUSSION In this study, we employed high-resolution single-cell transcriptomic sequencing to comprehensively investigate the tumor microenvironment associated with lymph node metastasis in penile cancer (PC). Unlike previous studies that primarily focused on changes in cellular composition or abundance ( 11 ) ( 12 ) , our clustering analysis revealed no statistically significant differences in the proportions of major cell types between lymph node metastasis–positive and –negative groups. This finding suggests that the key drivers of penile cancer metastasis may not simply depend on quantitative changes in specific cell populations. Machine learning represents a powerful approach for sample classification and identification of key features influencing classification outcomes, and the integration of multiple machine learning algorithms can further enhance predictive accuracy ( 13 ) . Accordingly, we integrated Random Forest, LASSO regression, and XGBoost algorithms and identified PLA2G2A as a candidate gene by extracting feature importance and intersecting the results across these methods. PLA2G2A has been reported to exhibit context-dependent and heterogeneous roles across different tumor types. As a phospholipase involved in phospholipid metabolism, PLA2G2A and its major product, arachidonic acid (AA), have been implicated in multiple oncogenic processes ( 14 ) . Previous studies have shown that PLA2G2A is expressed in cancer-associated fibroblasts (CAFs) in breast cancer and pancreatic ductal adenocarcinoma, where it promotes immune cell infiltration ( 15 ) ( 16 ) . In contrast, high PLA2G2A expression has been associated with a significantly reduced risk of recurrence in small cell lung cancer (SCLC) and gastric cancer ( 17 ) ( 18 ) . In the present study, by integrating differential expression analysis and survival analysis, we demonstrated that PLA2G2A is significantly upregulated in lymph node metastasis–positive penile cancer and is positively correlated with lymph node metastasis and poor clinical outcomes. These findings support the identification of PLA2G2A as a key pro-metastatic gene in penile cancer. Single-cell transcriptomic profiling revealed that PLA2G2A was specifically expressed in a distinct subpopulation of cancer-associated fibroblasts (CAFs), which we defined as PLA2G2A⁺ CAFs. Although this subset shared certain transcriptional features with progenitor-like CAFs, we observed that it also uniquely expressed CXCL12, suggesting that it may exert its functions through a distinct immunoregulatory network within the penile cancer microenvironment. CAFs are known to recruit myeloid cells via the secretion of chemokines and to induce their immunosuppressive polarization, a finding consistent with the work of Hong Liu et al. ( 19 ) . Cell–cell communication analysis focusing on PLA2G2A⁺ CAFs demonstrated that this subpopulation exerts broad immune cell recruitment effects through the CXCL12–CXCR4 axis. More importantly, we found that PLA2G2A⁺ CAFs specifically interact with myeloid cells via the CSF1–CSF1R pathway, which plays a critical role in macrophage recruitment and the induction of macrophage exhaustion ( 20 ) . To further elucidate the impact of PLA2G2A⁺ CAFs on macrophages, we performed a refined subclustering of myeloid cells and identified monocytes, transitional macrophages, M1 macrophages, M2 macrophages, and proliferating macrophages. Ligand–receptor interaction analysis further revealed that PLA2G2A⁺ CAFs recruit non–tumor-associated macrophages (non-TAMs) via CXCL12, while the MDK–LRP1 and MDK–NCL signaling axes directly act on both M1 and M2 macrophages. These signaling pathways have been reported to effectively promote macrophage polarization from a pro-inflammatory M1 phenotype toward an anti-inflammatory, tumor-promoting M2 phenotype ( 21 ) ( 22 ) , strongly supporting the notion that PLA2G2A⁺ CAFs are key drivers of immunosuppressive phenotypes within the tumor microenvironment. We further conducted pseudotime trajectory analysis of the monocyte–macrophage population, which clearly delineated the differentiation trajectory from monocytes to macrophages and the polarization process from M1 to M2 macrophages. Cells at the beginning of the trajectory predominantly consisted of monocytes characterized by high expression of FCN1 and VCAN. As pseudotime progressed, these cells transitioned through intermediate states and ultimately differentiated into M2 macrophages with high expression of FOLR2, CD163, and TREM2. Notably, CSF1R expression exhibited an overall increasing trend along pseudotime, indicating progressive macrophage maturation within the tumor. Meanwhile, the receptors for MDK, LRP1 and NCL, maintained relatively high expression levels during the mid-to-late stages of pseudotime (corresponding to the M1-to-M2 transition), which is consistent with our conclusion that PLA2G2A⁺ CAFs drive macrophage polarization via the MDK–LRP1/NCL axis. In addition, cell–cell communication analysis revealed VEGFA–VEGFR1 signaling interactions between PLA2G2A⁺ CAFs and endothelial cells. The VEGFA–VEGFR1 axis is a classical angiogenic pathway that has been well documented to increase vascular permeability, promote tumor angiogenesis, and facilitate tumor cell intravasation. Activation of this pathway has also been closely associated with lymph node metastasis in multiple malignancies ( 23 ) . However, in the specific context of penile cancer, whether PLA2G2A⁺ CAFs directly remodel vascular or lymphatic networks through enhanced VEGFA secretion to mediate lymph node metastasis, as well as the precise downstream molecular mechanisms involved, remain incompletely understood and are rarely reported in the literature. Although the present study did not include in-depth molecular validation of this pathway, this bioinformatic finding suggests that PLA2G2A⁺ CAFs may possess vascular remodeling potential in addition to their immunomodulatory functions. These observations provide important clues and novel directions for future investigations into the complex mechanisms by which PLA2G2A⁺ CAFs cooperate with the vascular system to promote penile cancer metastasis. Although this study identified novel biomarkers for lymph node metastasis in penile cancer, several limitations remain. First, regarding data cohort and model validation, the relatively low incidence of penile cancer, limited sample size, and homogeneous data type inevitably introduce selection bias. More importantly, while we combined multiple machine learning algorithms to improve the accuracy of key feature (PLA2G2A) screening, the lack of large-scale penile cancer cohorts (e.g., large transcriptome or single-cell datasets) with complete follow-up information on lymph node metastasis in public databases prevented us from validating our predictions on independent, multi-center external datasets. Second, in terms of mechanism exploration and functional validation, our conclusions are primarily based on transcriptome-level data mining and algorithmic predictions. Although intercellular communication and pseudo-temporal analysis strongly suggest that PLA2G2A⁺ CAFs may drive macrophage M2 polarization through the MDK-LRP1/NCL axis, these findings mainly reflect associations in gene expression rather than direct causal relationships. How PLA2G2A, as a lipid metabolism enzyme, regulates the secretion of its downstream chemokines remains unclear. To overcome these limitations, our future research will proceed in two directions: First, we plan to systematically collect standardized clinical, pathological, imaging, and biological data from newly diagnosed penile cancer patients. Based on this expanded domestic cohort, we will conduct more rigorous clinical validation of the correlation between PLA2G2A and its associated microenvironmental biomarkers and lymph node metastasis to compensate for the lack of external independent datasets. Second, we will dedicate ourselves to conducting in-depth molecular biology experiments using in vitro cell co-culture systems and in vivo animal lymph node metastasis models to ultimately elucidate the precise biological functions of PLA2G2A⁺ CAFs and their associated signaling pathways in driving penile cancer lymph node metastasis. Abbreviations AA: Arachidonic Acid AUC: Area Under the Curve BMI: Body Mass Index CAFs: Cancer-Associated Fibroblasts CI: Confidence Interval DFS: Disease-Free Survival EAU: European Association of Urology ECM: Extracellular Matrix FFPE: Formalin-Fixed Paraffin-Embedded GEO: Gene Expression Omnibus GO: Gene Ontology HDL: High-Density Lipoprotein HIV: Human Immunodeficiency Virus HPV: Human Papillomavirus HR: Hazard Ratio iCAFs: Inflammatory CAFs IHC: Immunohistochemistry LASSO: Least Absolute Shrinkage and Selection Operator LDL: Low-Density Lipoprotein LNM: Lymph Node Metastasis logFC: Log Fold Change myCAFs: Myofibroblastic CAFs NCCN: National Comprehensive Cancer Network NK cells: Natural Killer cells OS: Overall Survival PC / PeCa: Penile Cancer PCA: Principal Component Analysis pHB: Proportions of Hemoglobin genes PLA2G2A: Phospholipase A2 Group IIA pMT: Proportions of Mitochondrial genes pRP: Proportions of Ribosomal genes PSCC: Penile Squamous Cell Carcinoma ROC: Receiver Operating Characteristic SCLC: Small Cell Lung Cancer SM cells: Smooth Muscle cells TAMs: Tumor-Associated Macrophages TCGA: The Cancer Genome Atlas TME: Tumor Microenvironment XGBoost: Extreme Gradient Boosting Declarations Acknowledgments We thank specimen donors of patients of Yunnan Cancer Hospital and research groups that make project data publicly available. We would also like to express our sincere gratitude to the team of Maarten Albersen for providing the rare single-cell data on penile cancer. Data availability The scRNA-seq dataset obtained from Maarten Albersen's group at University Hospitals Leuven, DOI 10.1016/j.eururo.2024.03.038. The data comes from single-cell data of penile cancer published in the European Genome-Phenome Archive (EGA),which is hosted by the EBI and the CRG, under accession number EGAS50000000217. Consent for publication statement All authors read and approved the final version of the manuscript. Disclosure Funding This study was supported by Scientific Research Fund of Yunnan Provincial Department of Education (no. 2026J0278) and the HengruiResearch Fund of Kunming Medical University (YQHR2025-M18). Conflict of interest disclosure The authors have no relevant conflicts to declare. Ethical statement Informed Consent: All patients signed informed consent forms. Registry and the Registration No. of the study/trial: This study was approved by the Medical Ethics Committee of Yunnan Cancer Hospital (Approval No. (KYLX2022051)). Authors contribution Chen Hu : study design, methodology, writing. Weijia Chen : data analysis, validation. Fan Li : methodology, validation. Minghong Feng : formal analysis. Jianglai Wang : literature search. Zijian Tan: investigation. Bai yu: project supervision, funding acquisition. Hong Shi : project supervision, funding acquisition. References Douglawi AM, Timothy A. Penile cancer epidemiology and risk factors: a contemporary review. Current opinion in urology. 2019;29(2). Olesen TB, Sand FL, Rasmussen CL, Albieri V, Toft BG, Norrild B, et al. Prevalence of human papillomavirus DNA and p16INK4a in penile cancer and penile intraepithelial neoplasia: a systematic review and meta-analysis. The Lancet Oncology. 2018. White B, Dreger D, White BT, Gaul JM. Penile cancer. Nature reviews Disease primers. 2021;7 1:12. Kieffer JM, Djajadiningrat RS, Van Muilekom EAM, Graafland NM, Horenblas S, Aaronson NK. Quality of Life for Patients Treated for Penile Cancer. Journal of Urology. 2014;192(4):1105-10. Djajadiningrat RS, Graafland NM, Van Werkhoven E, Meinhardt W, Bex A, Van dP, Henk G., et al. Contemporary Management of Regional Nodes in Penile Cancer—Improvement of Survival? Journal of Urology. 2014;191(1):68-73. Pagliaro LC, Crook J. Multimodality therapy in penile cancer: when and which treatments? World Journal of Urology. 2009;27(2):221-5. Woldu SLC, BoHutchinson, Ryan C.Krabbe, Laura-MariaSingla, NirmishPassoni, Niccolo M.Clinton, Timothy N.Raj, Ganesh V.Miller, David S.Sagalowsky, Arthur I.Lotan, YairXie, YangMargulis, VitalyBagrodia, Aditya. Usage and survival implications of surgical staging of inguinal lymph nodes in intermediate- to high-risk, clinical localized penile cancer: A propensity-score matched analysis. Urologic oncology. 2018;36(4). Jeanne-Julien A, Bouchot O, De Vergie S, Branchereau J, Perrouin-Verbe M-A, Rigaud J. Morbidity and risk factors for complications of inguinal lymph node dissection in penile cancer. World Journal of Urology. 2023;41(1):109-18. Kroon B, Horenblas S, Lont A, Tanis P, Gallee M, Nieweg O. Patients with penile carcinoma benefit from immediate resection of clinically occult lymph node metastases. Journal of Urology. 2005;173(3):816-9. Elst L, Philips G, Vandermaesen K, Bassez A, Lodi F, Vreeburg MT, et al. Single-cell atlas of penile cancer reveals TP53 mutations as a driver of an aggressive phenotype, irrespective of human papillomavirus status, and provides clues for treatment personalization. European Urology. 2024;86(2):114-27. Zhang J, Lu T, Lu S, Ma S, Han D, Zhang K, et al. Single-cell analysis of multiple cancer types reveals differences in endothelial cells between tumors and normal tissues. Computational and structural biotechnology journal. 2023;21:665-76. Zhang Q, Liu Y, Wang X, Zhang C, Hou M, Liu Y. Integration of single-cell RNA sequencing and bulk RNA transcriptome sequencing reveals a heterogeneous immune landscape and pivotal cell subpopulations associated with colorectal cancer prognosis. Frontiers in Immunology. 2023. Vasanthakumar GU, Riadhwsein R, Priyanka GB, Aishwarya B, Kaliappan S, editors. Identification of Lung Cancer Biomarkers from Single-Cell RNA Sequencing Data using Random Forest with Extreme Gradient Boosting. 2025 4th International Conference on Distributed Computing and Electrical Circuits and Electronics (ICDCECE). Hidalgo I, Sorolla MA, Sorolla A, Salud A, Parisi E. Secreted Phospholipases A2: Drivers of Inflammation and Cancer. International Journal of Molecular Sciences. 2024;25(22). Liu T, Liu C, Yan M, Zhang L, Zhang J, Xiao M, et al. Single cell profiling of primary and paired metastatic lymph node tumors in breast cancer patients. Nature Communications. Wang Y, Liang Y, Xu H, Zhang X, Mao T, Cui J, et al. Single-cell analysis of pancreatic ductal adenocarcinoma identifies a novel fibroblast subtype associated with poor prognosis but better immunotherapy response. Cell Discovery. 2021;7. Zhu H, Shi H, Lu J, Zhu K, Yang L, Guo L, et al. Proteomic profiling reveals the significance of lipid metabolism in small cell lung cancer recurrence and metastasis. Journal of Translational Medicine. 2024;22(1):1117. Xing XF, Li H, Zhong XY, Zhang LH, Wang XH, Liu YQ, et al. Phospholipase A2 group IIA expression correlates with prolonged survival in gastric cancer. Histopathology. 2011;59(2):198-206. Liu T, Wang Z, Xue X, Wang Z, Zhang Y, Mi Z, et al. Single-cell transcriptomics analysis of bullous pemphigoid unveils immune-stromal crosstalk in type 2 inflammatory disease. Nature Communications. 2024;15(1):5949. Wang Q, Wang J, Xu K, Luo Z. Targeting the CSF1/CSF1R signaling pathway: an innovative strategy for ultrasound combined with macrophage exhaustion in pancreatic cancer therapy. Frontiers in Immunology. 2024;15:1481247. Zhang Y, Zuo C, Liu L, Hu Y, Yang B, Qiu S, et al. Single-cell RNA-sequencing atlas reveals an MDK-dependent immunosuppressive environment in ErbB pathway-mutated gallbladder cancer. Journal of hepatology. 2021;75(5):1128-41. Liu T, Ding Q, Gou J, Lu C, Lu X, Chen J, et al. Single-cell and spatial transcriptomics integration reveals FAM49B promotes tumor-associated macrophages polarization in colorectal cancer via the MK pathway. Frontiers in Immunology. 2025;16:1682637. Liu W, Xu J, Wang M, Wang Q, Bi Y, Han M. Tumor-derived vascular endothelial growth factor (VEGF)-a facilitates tumor metastasis through the VEGF-VEGFR1 signaling pathway. International journal of oncology. 2011;39(5):1213-20. Tables Tables 1 and 2 are available in the Supplementary Files section. Additional Declarations No competing interests reported. Supplementary Files Table1.xls Table2.xls Supplyment.pdf Supplementary Figure S1. Analysis of cell proportions stratified by metastatic status. Stacked bar plot comparing the relative proportions of major cell types between the LNM- and LNM+ groups. No significant lineage-specific expansion was observed at the major cell type level. Figure S2. UMAP plot showing cell distribution colored by LNM status (positive vs. negative), indicating enrichment of the PLA2G2A+ sub-cluster in the positive group. Figure S3. Feature plot illustrating the expression distribution of the PLA2G2A gene across fibroblast sub-clusters. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8936508","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":618140206,"identity":"1c692109-5bda-4baa-9ea8-6028c21564c4","order_by":0,"name":"hu chen","email":"","orcid":"","institution":"Kunming University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"hu","middleName":"","lastName":"chen","suffix":""},{"id":618140207,"identity":"b74fe17c-0802-4e9d-978e-82cc0635dd9f","order_by":1,"name":"chen weijia","email":"","orcid":"","institution":"Kunming University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"chen","middleName":"","lastName":"weijia","suffix":""},{"id":618140208,"identity":"7d31fe5e-0d7d-434f-b7d2-61ea89dcc86b","order_by":2,"name":"Li fan","email":"","orcid":"","institution":"Kunming University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Li","middleName":"","lastName":"fan","suffix":""},{"id":618140209,"identity":"2841c825-b732-40c9-84c3-044a88892fb9","order_by":3,"name":"feng minghong","email":"","orcid":"","institution":"Kunming University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"feng","middleName":"","lastName":"minghong","suffix":""},{"id":618140211,"identity":"eb9fdb78-f5c6-40c0-9b9f-62aa9cc606d0","order_by":4,"name":"wang jianglai","email":"","orcid":"","institution":"Yunnan Cancer Hospital","correspondingAuthor":false,"prefix":"","firstName":"wang","middleName":"","lastName":"jianglai","suffix":""},{"id":618140213,"identity":"c6480876-ba45-42c6-b91f-f3da47fd048a","order_by":5,"name":"tan zijian","email":"","orcid":"","institution":"Yunnan Cancer Hospital","correspondingAuthor":false,"prefix":"","firstName":"tan","middleName":"","lastName":"zijian","suffix":""},{"id":618140214,"identity":"4e706abd-0cd0-447f-ab8c-6cb33d68bb7e","order_by":6,"name":"bai yu","email":"","orcid":"","institution":"Yunnan Cancer Hospital","correspondingAuthor":false,"prefix":"","firstName":"bai","middleName":"","lastName":"yu","suffix":""},{"id":618140215,"identity":"0f317488-6f99-460f-8491-fd95abb83264","order_by":7,"name":"shi hong","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA3ElEQVRIiWNgGAWjYBACxgbGhgMJPyR42NjbDz5IqKghRgvzwQMfe2xk+HjOJBs8OHOMGHvYkg/OYEuzkZNIMJN82MJMWAPzjByDwzw8h3nYGBLSKhIb2Bj427sT8DsMrMUCpOXgsRuJO2QYJM6c3UCEFpAtjA1pNxLPsDEYSOQSo4UNiJgZzAoS25iJ0ZKWAPI+DxsbgxkDcVp6Hh8ABTIPGw9PskTCmWM8BP1i2J7Y/AEYlfby858f/PijokaOv72XgJYJCagCPHiVg4A8/wGCakbBKBgFo2CkAwAsHkzi1V9djwAAAABJRU5ErkJggg==","orcid":"","institution":"Kunming University of Science and Technology","correspondingAuthor":true,"prefix":"","firstName":"shi","middleName":"","lastName":"hong","suffix":""}],"badges":[],"createdAt":"2026-02-22 02:38:58","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8936508/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8936508/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":106703176,"identity":"03e098e2-722d-4ba4-9040-8fb9ab8f8459","added_by":"auto","created_at":"2026-04-12 07:39:34","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":2224280,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eThe Single-Cell Transcriptomic Atlas of Penile Cancer.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) A UMAP representation depicting 27 unique cell clusters derived from a total of 60,631 individual cells.\u003c/p\u003e\n\u003cp\u003e(B) A UMAP illustration highlighting 10 primary cell lineages, categorized according to the presence of specific marker genes.\u003c/p\u003e\n\u003cp\u003e(C) A UMAP diagram differentiated by patient identity, showcasing the cell distribution among samples and the success of batch effect mitigation.\u003c/p\u003e\n\u003cp\u003e(D) A UMAP chart indicating lymph node metastasis (LNM) status, with red representing negative cases and gold for positive ones.\u003c/p\u003e\n\u003cp\u003e(E) A dot plot that reveals the expression levels of key marker genes for each identified cell type, where dot size reflects the proportion of cells expressing the gene and color intensity indicates the average expression level.\u003c/p\u003e\n\u003cp\u003e(F) A stacked bar chart that represents the cellular makeup and relative proportions of major cell types across all samples.\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-8936508/v1/2e64d3bbb93eb7793a9a6a75.png"},{"id":106703179,"identity":"ec475f32-1578-4344-be43-b31cdb4ec6b5","added_by":"auto","created_at":"2026-04-12 07:39:34","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1286610,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eIdentification of key genes linked to lymph node metastasis through machine learning techniques.\u003cbr\u003e\n \u003c/strong\u003e(A) ROC curves assessing the effectiveness of LASSO, Random Forest, and XGBoost models, with AUC values reflecting each model's predictive precision.\u003c/p\u003e\n\u003cp\u003e(B) Confusion matrices that demonstrate the classification results for the three machine learning approaches.\u003c/p\u003e\n\u003cp\u003e(C) Bar charts ranking the top 20 feature genes determined by LASSO, Random Forest, and XGBoost according to their importance.\u003c/p\u003e\n\u003cp\u003e(D) A Venn diagram showing the commonality among the top 20 candidate genes identified by the three methods, with PLA2G2A being the only gene found in all.\u003c/p\u003e\n\u003cp\u003e(E) A volcano plot illustrating the differentially expressed genes between the metastasis-positive and negative groups in single-cell analysis, highlighting that PLA2G2A is significantly elevated in the positive cohort.\u003c/p\u003e\n\u003cp\u003e(F) A violin plot contrasting PLA2G2A expression levels between LNM- and LNM+ groups (****p \u0026lt; 0.0001).\u003c/p\u003e\n\u003cp\u003e(G) A volcano plot illustrating the genes that are expressed differently between the groups with and without metastasis in our BULK dataset. The gene PLA2G2A shows a notable increase in expression within the metastasis-positive group.\u003c/p\u003e\n\u003cp\u003e(H) A feature plot that displays how PLA2G2A expression is distributed across the UMAP embedding, with the expression primarily concentrated in the fibroblast cluster.\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-8936508/v1/9f73ccae4007acb5f0f3e6ef.png"},{"id":106703214,"identity":"5d37c480-70de-48a7-8c45-fce9076bdf58","added_by":"auto","created_at":"2026-04-12 07:39:47","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":15364996,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eVariation in PLA2G2A Expression Between Clinical Samples with Negative and Positive Lymph Node Status\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-8936508/v1/2823d6cf8e97d55f87543a97.png"},{"id":106703199,"identity":"1ba70f1c-f3ff-48e6-ad0d-2014af02d875","added_by":"auto","created_at":"2026-04-12 07:39:42","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":683816,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eAssociation Between PLA2G2A Expression and Clinical Prognosis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) DFS comparison between groups with positive and negative PLA2G2A; (B) OS analysis in groups categorized by positive and negative PLA2G2A.\u003c/p\u003e","description":"","filename":"Figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-8936508/v1/5edac33b20e99140e59b3436.png"},{"id":106703150,"identity":"5fc28d4a-377b-4263-81c4-0fbc7fea3485","added_by":"auto","created_at":"2026-04-12 07:39:30","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":1241019,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eIdentification and functional enrichment of the PLA2G2A+ CAF sub-cluster in penile cancer.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(A) \u003c/strong\u003eUMAP embedding showing the 14 distinct clusters identified upon sub-clustering of the fibroblast lineage.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(B)\u003c/strong\u003e The six identified fibroblast subtypes are depicted using UMAP visualization, with the PLA2G2A+ CAF group distinctly marked in blue.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(C) \u003c/strong\u003eA dot plot illustrating key marker genes for various fibroblast subtypes. The PLA2G2A+ group is notably defined by elevated levels of PLA2G2A and CXCL12.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(D)\u003c/strong\u003e A dot plot showcasing the biological processes from Gene Ontology (GO) that are notably enriched in the CAF subpopulation characterized by PLA2G2A.\u003c/p\u003e","description":"","filename":"Figure5.png","url":"https://assets-eu.researchsquare.com/files/rs-8936508/v1/fe8c8cda9bc9c7ad6ba83480.png"},{"id":106703219,"identity":"06f05981-0a80-4648-99be-cf3969d99454","added_by":"auto","created_at":"2026-04-12 07:39:48","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":1901171,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eInteractions Between PLA2G2A⁺ CAFs and Other Cellular Subpopulations\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eA)\u003c/strong\u003e A chord diagram that represents the overall strength of communication between PLA2G2A⁺ CAFs and the primary cell types found in the tumor microenvironment (TME), with line thickness indicating the weight of interactions.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(B) \u003c/strong\u003eA dot plot that showcases significant ligand-receptor interactions from PLA2G2A⁺ CAFs to various other cell types, where color intensity reflects the likelihood of communication and the size of the dots represents statistical significance (P-value).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(C)\u003c/strong\u003e A UMAP representation of myeloid cell subpopulations, identifying eight unique clusters, including monocytes, M1/M2 macrophages, and transitional tumor-associated macrophages (TAMs).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(D)\u003c/strong\u003e A dot plot that illustrates the expression patterns of key marker genes for each myeloid subpopulation.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(E)\u003c/strong\u003e A dot plot that emphasizes particular ligand-receptor interactions between PLA2G2A⁺ CAFs and specific myeloid subsets.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(F, G)\u003c/strong\u003eChord diagrams that illustrate the structures of intercellular signaling networks for the CXCL (F) and MK (MDK) (G) pathways.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(H)\u003c/strong\u003e A heatmap that describes the signaling functions (sender, receiver, mediator, influencer) of various cell types within the CXCL and MK signaling networks.\u003c/p\u003e","description":"","filename":"Figure6.png","url":"https://assets-eu.researchsquare.com/files/rs-8936508/v1/6d5c33e41609d06dd8e7423a.png"},{"id":106728269,"identity":"4225be9c-8d89-4a9c-931f-d028be1caf20","added_by":"auto","created_at":"2026-04-12 18:42:19","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":1887594,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePseudotime trajectory analysis delineates the monocyte-to-macrophage differentiation lineage and associated transcriptional dynamics.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(A)\u003c/strong\u003e Monocle 3 trajectory inference of myeloid cells, colored by cell type (left) and pseudotime (right). The trajectory reveals a developmental continuum initiating from \u003cstrong\u003emonocytes (root)\u003c/strong\u003e, progressing through an \u003cstrong\u003eM1/intermediate state\u003c/strong\u003e, and terminating at \u003cstrong\u003eM2 macrophages (terminal state)\u003c/strong\u003e.\u003cbr\u003e\n \u003cstrong\u003e(B)\u003c/strong\u003e Temporal expression kinetics of key marker genes along the pseudotime axis. Solid black lines indicate smoothed fitted expression trends.\u003cbr\u003e\n \u003cstrong\u003e(C)\u003c/strong\u003e UMAP feature plots visualizing the spatial expression patterns of the corresponding genes.\u003c/p\u003e","description":"","filename":"Figure7.png","url":"https://assets-eu.researchsquare.com/files/rs-8936508/v1/9363ef28214358d1aa2ee1e8.png"},{"id":109220235,"identity":"6b5aa894-e530-4ff0-bb25-e67db7065426","added_by":"auto","created_at":"2026-05-13 20:13:04","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":22458935,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8936508/v1/db7027a9-21e2-4340-abe2-458968ecacb0.pdf"},{"id":106703148,"identity":"40f69c22-82cc-4dbd-bfa1-2ca334522804","added_by":"auto","created_at":"2026-04-12 07:39:29","extension":"xls","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":25600,"visible":true,"origin":"","legend":"","description":"","filename":"Table1.xls","url":"https://assets-eu.researchsquare.com/files/rs-8936508/v1/ba446be5b3aa85d92ba9268a.xls"},{"id":106703153,"identity":"9c1ed773-d5d2-41f2-a02f-40ff2844b1d6","added_by":"auto","created_at":"2026-04-12 07:39:30","extension":"xls","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":22528,"visible":true,"origin":"","legend":"","description":"","filename":"Table2.xls","url":"https://assets-eu.researchsquare.com/files/rs-8936508/v1/c042626ca3493bfd865b7d95.xls"},{"id":106703149,"identity":"6bfad26f-aa7d-4ac8-9afd-d97888b7e5b0","added_by":"auto","created_at":"2026-04-12 07:39:30","extension":"pdf","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":558480,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSupplementary\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFigure S1. Analysis of cell proportions stratified by metastatic status.\u003c/strong\u003e\u003cbr\u003e\nStacked bar plot comparing the relative proportions of major cell types between the LNM- and LNM+ groups. No significant lineage-specific expansion was observed at the major cell type level.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFigure S2.\u003c/strong\u003e UMAP plot showing cell distribution colored by LNM status (positive vs. negative), indicating enrichment of the PLA2G2A+ sub-cluster in the positive group.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFigure S3. \u003c/strong\u003eFeature plot illustrating the expression distribution of the PLA2G2A gene across fibroblast sub-clusters.\u003c/p\u003e","description":"","filename":"Supplyment.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8936508/v1/07f0a8e18f25d1033125c123.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"PLA2G2A⁺ CAFs as a Key Driver of Lymph Node Metastasis in Penile Cancer and Its Underlying Mechanisms","fulltext":[{"header":"INTRODUCTION ","content":"\u003cp\u003ePenile cancer is an uncommon type of cancer, exhibiting significant differences in occurrence among various global populations. In some areas of Africa, Asia, and South America, it can represent as much as 10% of all cancers affecting men\u003csup\u003e(1, 2)\u003c/sup\u003e. In China, the current rate of penile cancer is estimated to be around 0.5 to 2.5 cases per 100,000 individuals\u003cstrong\u003e\u003csup\u003e(3)\u003c/sup\u003e\u003c/strong\u003e. While the overall frequency is low, advanced stages of this disease can be life-threatening and greatly diminish the quality of life for those affected. Treatments for penile cancer frequently lead to considerable anatomical and functional impairments in the genital region, resulting in issues such as urinary and sexual dysfunction, along with considerable psychological distress due to changes in appearance.\u003csup\u003e(4)\u003c/sup\u003e. Consequently, even though it is a rare form of cancer, penile cancer deserves increased public awareness, with a focus on improving treatment options and enhancing the long-term well-being of patients.\u003c/p\u003e\n\u003cp\u003ePenile cancer is an uncommon type of cancer, especially infrequent in advanced Western nations. This rarity has led to insufficient extensive and high-caliber foundational studies, along with considerable obstacles in executing randomized phase III clinical trials. As a result, there are notable deficiencies in both translational and basic research within this area.\u003c/p\u003e\n\u003cp\u003eYunnan Province in China, known for its economic challenges, experiences a notably high rate of penile cancer. This situation is linked to various risk factors, such as a significant number of smokers, inadequate hygiene that causes chronic inflammation due to smegma buildup, and a widespread occurrence of HIV, which correlates with higher HPV infection rates. Together, these elements raise the likelihood of penile cancer in the area. Consequently, the increased prevalence of this disease in Yunnan creates advantageous and concrete opportunities for conducting penile cancer research at our facility.\u003c/p\u003e\n\u003cp\u003eThe prognosis for penile cancer shows significant variation. For men with localized penile cancer (N0 stage), the 5-year survival rate is around 90%\u003csup\u003e(5)\u003c/sup\u003e . However, the situation worsens with the onset of lymph node metastasis. Patients with unilateral inguinal lymph node involvement of two or fewer nodes (N1 stage) have a 5-year survival rate of about 80%, but this figure plummets to between 10% and 20% for those with bilateral inguinal or pelvic lymph node involvement (N2 or N3 stage). In instances where there is extranodal lymph node spread, the overall survival rate drops below 10%\u003csup\u003e(6)\u003c/sup\u003e. Consequently, the presence of multiple lymph node metastases (indicative of advanced penile cancer) leads to a significant decline in patient survival rates.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTimely identification and intervention are essential factors influencing patient survival rates, with lymph node metastasis being significantly linked to unfavorable outcomes. Research indicates that performing early inguinal lymph node dissection can improve survival chances\u003csup\u003e(7)\u003c/sup\u003e. Nonetheless, this procedure often results in postoperative issues, particularly lymphatic leakage, which may require extended drainage periods. Furthermore, patients may experience complications like lymphedema in the lower limbs, infections or necrosis of the surgical site, and hernia development, all of which can adversely affect their long-term quality of life\u003csup\u003e(8)\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eCurrently, guidelines from the NCCN, EAU, and Chinese authorities advocate for immediate bilateral inguinal lymph node dissection when enlarged inguinal lymph nodes are detected, irrespective of whether penile cancer is categorized as low, intermediate, or high risk. Notably, around 30% of intermediate-risk patients without clinically evident lymphadenopathy are discovered to have hidden micrometastases upon postoperative pathological assessment\u003csup\u003e(9)\u003c/sup\u003e. Conversely, some patients exhibiting palpable inguinal lymphadenopathy do not show any metastatic involvement in their postoperative pathology results. This indicates a lack of sufficiently sensitive biomarkers or imaging techniques to reliably predict the existence of hidden micrometastases in the inguinal lymph nodes. Consequently, certain patients may experience overtreatment, leading to avoidable complications, while others may not receive adequate treatment, resulting in disease advancement. Thus, determining the right candidates for inguinal lymph node dissection in early-stage penile cancer is a pressing and unresolved clinical issue. This study seeks to combine single-cell transcriptomic analysis with machine learning methods to pinpoint molecular markers that could aid in the preoperative assessment of lymph node metastasis. These markers could enhance clinical decision-making regarding the necessity of inguinal lymph node dissection, helping to prevent overtreatment and its complications, as well as avoiding disease progression due to delayed or missed surgical interventions.\u003c/p\u003e"},{"header":"MATERIALS AND METHODS","content":"\u003cp\u003e\u003cstrong\u003eSingle-cell RNA-seq Data Acquisition and Construction of Atlas\u003c/strong\u003e\u003cbr\u003e\u0026nbsp;The initial raw data for single-cell RNA sequencing (scRNA-seq) was supplied by Maarten Albersen's group at University Hospitals Leuven\u003csup\u003e(10)\u003c/sup\u003e , aligning with earlier reports. Transcriptomic data along with clinical details for patients with esophageal cancer were retrieved from The Cancer Genome Atlas (TCGA). For external validation, additional cohorts were acquired from the Gene Expression Omnibus (GEO).\u003c/p\u003e\n\u003cp\u003eThe development of the Single-cell Atlas involved data analysis through the Seurat R package (version 4.0). Initial Seurat objects were created by applying filtering criteria that removed genes found in fewer than three cells and cells with fewer than 200 expressed features. After merging the datasets, quality control was conducted by evaluating the proportions of mitochondrial (pMT), hemoglobin (pHB), and ribosomal (pRP) genes. The data underwent normalization using the NormalizeData function, and the top 2,000 most variable features were selected. A linear transformation was executed with ScaleData (Z-score scaling), and batch effects were adjusted using the Harmony algorithm (RunHarmony). Dimensionality reduction was achieved through PCA, selecting the top 30 principal components for UMAP embedding. Cell clusters were determined using FindNeighbors and FindClusters (resolution set to 0.5). Differential expression analysis was performed with FindAllMarkers (logFC \u0026gt; 0.25, only.pos = TRUE), and cell types were annotated according to established markers from existing literature and public databases.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSample Population and Ethics Statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis forward-looking investigation involved individuals diagnosed with PSCC at Yunnan Cancer Hospital in Kunming, China, from September 2022 to April 2025. Out of 137 participants who satisfied the criteria for inclusion and exclusion, 77 had comprehensive follow-up information, while 69 had accessible formalin-fixed paraffin-embedded (FFPE) tissue samples. Informed written consent was obtained from these 69 patients, and the Medical Ethics Committee of Yunnan Cancer Hospital approved the study protocol (KYLX2022051) prior to the commencement of data collection. The research adhered to the principles outlined in the Declaration of Helsinki (2013 revision).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eInclusion criteria are as follows:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e1) Diagnosed with PSCC by pathology;\u003c/p\u003e\n\u003cp\u003e2) Willing to participate in the research protocol and willing to follow up.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eThe exclusion criteria are as follows:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e1) non-squamous cell carcinoma or the presence of other histological differentiation;\u003c/p\u003e\n\u003cp\u003e2) no lymph node dissection performed;\u003c/p\u003e\n\u003cp\u003e3) Lack of complete FFPE tissue blocks;\u003c/p\u003e\n\u003cp\u003e4) Patients with incomplete medical records or unable to obtain required clinical information;\u003c/p\u003e\n\u003cp\u003e5) Patients unwilling to provide clinical information.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eImmunohistochemistry (IHC) and Scoring Analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eImmunohistochemical analysis was conducted on tissue samples from 44 patients without lymph node involvement and 25 patients with lymph node metastasis. The IHC analysis utilized an antibody targeting PLA2G2A (1:200; Cat# PA5-102403, Thermo Fisher Scientific). Tissue slides were sourced from Yunnan Cancer Hospital, with collection occurring between November 2023 and August 2024. Informed consent was obtained from all participants, and the research protocol received approval from the Medical Ethics Committee at Yunnan Cancer Hospital prior to the commencement of data gathering.\u003c/p\u003e\n\u003cp\u003eMicroscopic images of sections were captured from five randomly selected fields. The protein levels were evaluated using the IHC profiler plugin in ImageJ. The scoring for the percentage of positively stained cells was based on these criteria: 0 for less than 10% positive cells, 1 for 10–25%, 2 for 26–50%, 3 for 51–75%, and 4 for more than 75%. Staining intensity was rated as follows: 0 for no staining, 1 for weak, 2 for moderate, and 3 for strong staining. The overall staining score was determined by multiplying the intensity score by the proportion score, yielding a total that could range from 0 to 12.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSurvival analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe analysis employing Kaplan-Meier was conducted with the R packages ‘survival’ (version 3.1–8) and visualized through ‘survminer’ (version 0.4.9). A detection threshold of 95% specificity was applied, with the exception of overall and recurrence-free survival calculations, which utilized a threshold of 99% specificity. The survival curves were derived from the overall survival data of patients, adhering to the 99% specificity threshold.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStatistical Analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eData collection and analysis were performed utilizing SPSS version 26.0 (IBM Corp., Armonk, NY, USA) and R version 4.2.1 (R Core Team, R Foundation for Statistical Computing, Vienna, Austria). The lymph node count is presented as mean ± standard deviation, and group comparisons were conducted using the t-test. A p-value of less than 0.05 was deemed statistically significant.\u003c/p\u003e"},{"header":"RESULTS","content":"\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eConstruction of the Single-Cell Transcriptomic Atlas for Penile Cancer\u003c/h2\u003e \u003cp\u003e In order to thoroughly explore the diversity within the tumor microenvironment (TME) of penile cancer (PeCa) and identify potential changes linked to lymph node metastasis (LNM), we examined a single-cell RNA sequencing dataset provided by Maarten Albersen's team at University Hospitals Leuven. The study included 17 patients, divided into two groups: 8 without lymph node metastasis (LNM-) and 9 with lymph node metastasis (LNM+). After applying stringent quality control measures, we selected 60,631 high-quality single cells for further analysis. The application of Uniform Manifold Approximation and Projection (UMAP) for dimensionality reduction resulted in the classification of these cells into 27 unique clusters \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA\u003cb\u003e)\u003c/b\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eUtilizing the expression profiles of key marker genes \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eE\u003cb\u003e)\u003c/b\u003e, the identified clusters were classified into ten primary cell types \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB\u003cb\u003e)\u003c/b\u003e. These classifications encompassed epithelial cells (characterized by KRT5hi, TP63hi, SFNhi), T lymphocytes (marked by CD3Dhi, CD3Ehi), fibroblasts (identified by COL1A1, DCN, LUM), myeloid cells (indicated by LYZ, CD68), endothelial cells (noted by PECAM1, VWF), natural killer (NK) cells (defined by GNLY, NKG7), B lymphocytes (identified by CD79A, MS4A1), plasma cells (characterized by IGHG1, JCHAIN), mast cells (marked by TPSB2, KIT), and smooth muscle (SM) cells (indicated by ACTA2, TAGLN).\u003c/p\u003e \u003cp\u003eTo validate the strength of our analysis, we assessed possible batch influences among the samples. The UMAP visualization revealed a consistent cell distribution throughout all clusters, regardless of the patient's origin \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eC\u003cb\u003e)\u003c/b\u003e or metastatic condition (LNM- vs. LNM+) \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eD\u003cb\u003e)\u003c/b\u003e, suggesting successful data integration with minimal bias related to batch effects.\u003c/p\u003e \u003cp\u003eIn conclusion, we examined the cellular makeup of the PeCa environment. Epithelial cells, T cells, and fibroblasts emerged as the main cell types present in the tumor microenvironment (TME) \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eF\u003cb\u003e)\u003c/b\u003e. Importantly, a comparison between the LNM- and LNM+ cohorts showed no significant differences in the relative proportions of the key cell types \u003cb\u003e(Supplementary Fig.\u0026nbsp;1)\u003c/b\u003e. This indicates that the process of lymph node metastasis in penile cancer may not be primarily influenced by the proliferation of certain cell lineages, but instead by changes in the functional states and inherent molecular traits of particular subpopulations. This insight calls for a more thorough exploration of the molecular features at the level of specific cell subsets.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eIdentification of PLA2G2A as a Predictive Marker for Lymph Node Metastasis by Machine Learning\u003c/h2\u003e \u003cp\u003eIn order to identify strong gene features linked to lymph node metastasis (LNM), we utilized three distinct machine learning techniques: LASSO, Random Forest, and XGBoost. The analysis of Receiver Operating Characteristic (ROC) curves revealed excellent predictive capabilities for each model, with Area Under the Curve (AUC) scores of 0.960, 0.931, and 0.962, respectively \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA\u003cb\u003e)\u003c/b\u003e. Additionally, confusion matrices validated the high classification precision of these approaches \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB\u003cb\u003e)\u003c/b\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFollowing this, we identified the 20 most significant genes from each method \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC\u003cb\u003e)\u003c/b\u003e. To determine the most dependable biomarker, we examined the overlap among these three gene collections. The Venn diagram analysis indicated that PLA2G2A was the only gene consistently found in the highest-ranked lists across all three methods \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD\u003cb\u003e)\u003c/b\u003e.\u003c/p\u003e \u003cp\u003eIn order to confirm this result, we analyzed gene expression patterns in both metastatic and non-metastatic samples. The volcano plot analysis revealed a notable increase in PLA2G2A levels within the LNM+ group \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eE\u003cb\u003e)\u003c/b\u003e. This finding was additionally supported by violin plots, which demonstrated a marked rise in PLA2G2A expression in metastatic samples when compared to LNM- controls \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eF\u003cb\u003e)\u003c/b\u003e.\u003c/p\u003e \u003cp\u003eTo ensure independent validation, we conducted bulk RNA sequencing on an additional group of four patients with penile cancer (two with lymph node metastasis and two without) from Yunnan Cancer Hospital. In line with our findings from single-cell analysis, the differential expression assessment revealed a notable increase in PLA2G2A levels in the group with lymph node metastasis \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eG\u003cb\u003e)\u003c/b\u003e.\u003c/p\u003e \u003cp\u003eIn conclusion, we charted the expression levels of PLA2G2A across our single-cell atlas. The feature plots indicated that PLA2G2A was mainly confined to fibroblast clusters, showing minimal expression in other cell types \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eH\u003cb\u003e)\u003c/b\u003e. This unique distribution underscores the significance of the stromal compartment, leading us to concentrate our further research on the various fibroblast subpopulations.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003ePLA2G2A are predictive markers for lymph node metastasis\u003c/h3\u003e\n\u003cp\u003eA cohort of 69 patients diagnosed with PSCC underwent immunohistochemistry (IHC) analysis. Comprehensive baseline information is presented in \u003cb\u003eTable\u0026nbsp;1\u003c/b\u003e. There were no notable differences in age, smoking habits, body mass index (BMI), tumor morphology, lesion site, size, presence of distant metastases, type of surgical intervention, tumor grade, triglyceride levels, total cholesterol, HDL cholesterol, LDL cholesterol, fibrinogen, or squamous cell carcinoma antigen between the groups with and without lymph node involvement. However, significant variations were noted in pathological T stage, overall survival, and recurrence rates between the two groups. The expression of PLA2G2A was significantly elevated in the lymph node-positive group (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05) \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e\u003cb\u003e)\u003c/b\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\n\u003ch3\u003eSpearman correlation analysis of different clinical pathological features between the N0 and N1 groups\u003c/h3\u003e\n\u003cp\u003eThrough the analysis of the relationships among the clinical characteristics mentioned earlier, it was initially determined that the pathological T stage, vital status, and recurrence status were linked to lymph node metastasis. In contrast, the other clinical features did not exhibit any notable correlations. Comprehensive findings can be found in \u003cb\u003eTable\u0026nbsp;2.\u003c/b\u003e\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eCorrelation of PLA2G2A With Clinical Outcomes\u003c/h2\u003e \u003cp\u003eWe categorized cases with IHC scores exceeding 0 for PLA2G2A as the positive cohort (P+), while those with an IHC score of 0 were designated as the negative cohort (P\u0026minus;). Among the 69 patients followed for a median duration of 41 months (interquartile range: 12.6\u0026ndash;62.8 months), 14 had recurrences and 13 succumbed. The Kaplan-Meier survival analysis indicated that positivity for PLA2G2A did not correlate with overall survival (hazard ratio [HR]\u0026thinsp;=\u0026thinsp;1.21; 95% CI: 0.4026\u0026ndash;3.637, p\u0026thinsp;\u0026gt;\u0026thinsp;0.05). However, it was linked to poorer disease-free survival (DFS) (hazard ratio [HR]\u0026thinsp;=\u0026thinsp;3.332; 95% CI: 1.156\u0026ndash;9.605, p\u0026thinsp;=\u0026thinsp;0.0259) \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e\u003cb\u003e).\u003c/b\u003e\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eIdentification of a PLA2G2A\u0026thinsp;+\u0026thinsp;CAF Fibroblast Subpopulation Associated with Lymph Node Metastasis\u003c/h2\u003e \u003cp\u003eTo explore the precise function of PLA2G2A in the lymph node metastasis of penile cancer, we conducted a detailed sub-clustering analysis of the fibroblast population. Through unsupervised clustering, we identified 14 unique initial clusters of fibroblasts \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA\u003cb\u003e)\u003c/b\u003e. These clusters were subsequently classified into six primary functional subtypes based on the expression patterns of key marker genes: PLA2G2A+ cancer-associated fibroblasts (CAFs), inflammatory CAFs (iCAFs), myofibroblastic CAFs (myCAFs), progenitor-like fibroblasts, tissue-specific fibroblasts, and various other types \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB\u003cb\u003e)\u003c/b\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eAn examination of the marker genes within each subcluster indicates that the PLA2G2A+ subpopulation is distinguished not only by the notable increase in PLA2G2A expression but also by elevated levels of the chemokine CXCL12 and ADH1B \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC\u003cb\u003e)\u003c/b\u003e. We then categorized the fibroblast atlas based on metastatic status. UMAP visualization revealed a significant concentration of the PLA2G2A+ subpopulation in the LNM+ cohort \u003cb\u003e(Supplementary Fig.\u0026nbsp;2)\u003c/b\u003e. Additionally, the spatial pattern of PLA2G2A expression closely corresponded with this cluster associated with metastasis \u003cb\u003e(Supplementary Fig.\u0026nbsp;3)\u003c/b\u003e, further emphasizing the strong link between this subpopulation and the metastatic characteristics.\u003c/p\u003e \u003cp\u003eTo clarify the functional characteristics of the PLA2G2A+ subpopulation, we conducted a Gene Ontology (GO) enrichment analysis on its defining genes. The results indicated a notable enrichment in pathways associated with the organization of the extracellular matrix (ECM), the structuring of extracellular components, and the proliferation of epithelial cells \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eD\u003cb\u003e)\u003c/b\u003e. This subpopulation demonstrated not only strong capabilities for matrix remodeling but also elevated levels of various secretory factors, such as CXCL12. This pronounced secretory function aligns with the identified enrichment in the \\response to endoplasmic reticulum stress\\ pathway. These results imply that PLA2G2A+ CAFs could facilitate lymph node metastasis by creating a specialized physical environment or by releasing chemokines that attract and reprogram immune cells, including macrophages, within the tumor microenvironment.\u003c/p\u003e \u003cp\u003e \u003cb\u003eCAF-derived PLA2G2A facilitates the attraction of monocytes through CXCL12 and promotes the transition from M1 to M2 macrophages via the MDK-NCL/LRP1 pathway.\u003c/b\u003e \u003c/p\u003e \u003cp\u003eTo thoroughly analyze how PLA2G2A+CAFs alter the tumor microenvironment (TME) through intercellular interactions\u0026mdash;particularly regarding the myeloid compartment\u0026mdash;we conducted an extensive ligand-receptor interaction study utilizing CellChat. We initially identified PLA2G2A+ CAFs as the signaling origin to chart their overall communication dynamics. The circle plot visualization indicated that the most significant interaction strengths were observed between PLA2G2A+ CAFs and various lymphocytes (including T cells, B cells, and NK cells) as well as myeloid cells \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA\u003cb\u003e)\u003c/b\u003e. At the ligand-receptor interaction level, the CXCL12-CXCR4 pathway was widely recognized between CAFs and multiple immune cell types, indicating that PLA2G2A+ CAFs have a broad chemotactic influence \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB\u003cb\u003e)\u003c/b\u003e. Notably, we detected a pronounced and specific enrichment of the CSF1-CSF1R pathway in interactions between CAFs and myeloid cells \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB\u003cb\u003e)\u003c/b\u003e. Since CSF1 is crucial for the survival, proliferation, and differentiation of macrophages, this observation highlights that PLA2G2A+ CAFs not only attract immune cells but also create a specialized ecological niche that fosters the survival and presence of myeloid populations.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTo clarify the unique effects of signals from cancer-associated fibroblasts (CAFs) on various myeloid differentiation states, we isolated and re-categorized myeloid cells into eight distinct sub-groups \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eC\u003cb\u003e)\u003c/b\u003e. Utilizing established marker profiles \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eD\u003cb\u003e)\u003c/b\u003e, we identified these groups as: Monocytes (FCN1+, VCAN+), M1 Macrophages (CXCL9+, CXCL10+, IL1B+), M2 Macrophages (CD163+, FOLR2+, TREM2+), Transitional Tumor-Associated Macrophages (TAMs) (C1QC+, APOE+, TMSB4X+), Proliferating TAMs (MKI67+, TOP2A+), along with other myeloid elements such as Neutrophils (CSF3R+, FCGR3B+), Langerhans Cells (CD207+, CD1A+), and Dendritic Cells (identified by cDC/pDC markers). This detailed classification established a framework for understanding the specific spatiotemporal dynamics of CAF signaling.\u003c/p\u003e \u003cp\u003eBy examining the interactions between PLA2G2A+ CAFs and various myeloid subpopulations \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eE\u003cb\u003e)\u003c/b\u003e, we mapped out a complex signaling network that encompasses both recruitment and polarization processes. During the recruitment stage, we found that the chemokine receptor CXCR4, which binds to CXCL12, was significantly present in non-TAM myeloid groups \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eE\u003cb\u003e)\u003c/b\u003e. Our pathway analysis identified PLA2G2A+ CAFs as the primary source of CXCL signals \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eF\u003cb\u003e)\u003c/b\u003e, suggesting that these CAFs mainly rely on CXCL12 to effectively draw precursor monocytes from the bloodstream into the tumor microenvironment, thus kickstarting immune infiltration.\u003c/p\u003e \u003cp\u003eOur findings on polarization revealed a crucial mechanism through which cancer-associated fibroblasts (CAFs) influence the phenotypic shift of macrophages. Midkine (MDK), a growth factor that binds heparin and is secreted, is recognized for its role in steering tumor-associated macrophages towards an immunosuppressive M2 phenotype by activating receptors like Nucleolin (NCL) and LRP1. Analysis of receptor expression indicated that M1 macrophages, which are generally linked to anti-tumor activity, exhibited a notable presence of the MDK receptors NCL and LRP1 \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eE\u003cb\u003e)\u003c/b\u003e. Additionally, CellChat analysis highlighted a significant signaling interaction involving MDK-NCL and MDK-LRP1 between PLA2G2A+ CAFs and M1 macrophages, with an intensity similar to that found in M2 macrophages \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eE\u003cb\u003e)\u003c/b\u003e. A comprehensive examination of the MK signaling pathway network \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eG\u003cb\u003e)\u003c/b\u003e and the heatmaps depicting signal roles \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eH\u003cb\u003e)\u003c/b\u003e validated that PLA2G2A+ CAFs are the main source of MDK signals, while both M1 and M2 macrophages act as crucial recipients.\u003c/p\u003e \u003cp\u003eThe gathered evidence indicates that PLA2G2A+ CAFs perform a synchronized two-phase regulatory role: first, they attract monocytes through CXCL12, and then they engage the NCL/LRP1 receptors on M1 macrophages through the MDK signaling pathway. This ongoing signaling is likely to trigger changes in gene expression that suppress the pro-inflammatory activities of M1 and encourage a shift towards the M2 phenotype, thus creating and maintaining an immunosuppressive environment that supports tumor advancement.\u003c/p\u003e \u003cp\u003e \u003cb\u003ePseudotime Trajectory Analysis Corroborates CAF-Driven Differentiation of Monocytes toward an M2-Like Phenotype via an M1-Like Intermediate\u003c/b\u003e \u003c/p\u003e \u003cp\u003eTo substantiate our hypothesis derived from ligand-receptor interaction analysis\u0026mdash;specifically, that PLA2G2A+ CAFs recruit monocytes and orchestrate their transition from an M1-like state toward an M2-like phenotype\u0026mdash;we performed pseudotime trajectory analysis on the identified monocyte and macrophage populations using Monocle 3. The inferred trajectory recapitulated a continuous developmental lineage, anchoring Monocytes at the root of the pseudo-temporal axis. As differentiation progressed, the trajectory revealed that monocytes evolved through an intermediate state composed of Transitional TAMs and M1 Macrophages, eventually culminating in terminal differentiation into M2 Macrophages, alongside a minor branch of Proliferating TAMs \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eA\u003cb\u003e)\u003c/b\u003e. This topological structure provides compelling computational evidence supporting a sequential differentiation model wherein recruited monocytes initially acquire M1-like characteristics but are ultimately reprogrammed into M2 macrophages within the tumor microenvironment.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTo dissect the molecular dynamics underlying this M1-to-M2 transition, we interrogated the dynamic expression profiles of canonical marker genes along the pseudo-time axis \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eB, \u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eC\u003cb\u003e).\u003c/b\u003e Aligning with the recruitment model, monocyte markers such as FCN1 and the CXCL12 receptor CXCR4 exhibited maximal expression at the trajectory root followed by a rapid decline, confirming that chemotactic responsiveness is a transient feature of early-stage monocytes. Crucially, analysis of polarization markers unveiled a dynamic \"phenotype switching\" event: the M1-associated proinflammatory chemokine CXCL9 displayed a distinct, transient surge during the intermediate phase (spanning Transitional TAMs and M1 Macrophages), followed by downregulation. In sharp contrast, M2-associated immunosuppressive markers\u0026mdash;including CD163, FOLR2, and TREM2\u0026mdash;exhibited progressive and sustained upregulation, peaking at the trajectory terminus. This pattern\u0026mdash;characterized by the transient induction and subsequent suppression of M1 features coupled with the progressive accumulation of M2 traits\u0026mdash;transcriptionally validates the reprogramming process from a proinflammatory to an immunosuppressive phenotype.\u003c/p\u003e \u003cp\u003eFurthermore, to validate the mechanistic drivers of this reprogramming, we mapped the expression of receptors for CAF-secreted ligands onto the trajectory. NCL (a receptor for MDK) maintained expression throughout the differentiation lineage, whereas LRP1 (another MDK receptor) and CSF1R (receptor for CSF1) showed a marked resurgence or elevation specifically at the terminal M2 stage (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB). The synchronized upregulation of these receptors with the acquisition of the M2 phenotype suggests that differentiating macrophages, under the regulatory influence of escalating MDK and CSF1 signals derived from PLA2G2A+ CAFs, undergo an initial differentiation toward an M1 stage before ultimately and stably transitioning into the M2 phenotype along the developmental trajectory.\u003c/p\u003e \u003c/div\u003e"},{"header":"DISCUSSION","content":"\u003cp\u003eIn this study, we employed high-resolution single-cell transcriptomic sequencing to comprehensively investigate the tumor microenvironment associated with lymph node metastasis in penile cancer (PC). Unlike previous studies that primarily focused on changes in cellular composition or abundance \u003csup\u003e(\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e) (\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e)\u003c/sup\u003e, our clustering analysis revealed no statistically significant differences in the proportions of major cell types between lymph node metastasis\u0026ndash;positive and \u0026ndash;negative groups. This finding suggests that the key drivers of penile cancer metastasis may not simply depend on quantitative changes in specific cell populations.\u003c/p\u003e \u003cp\u003eMachine learning represents a powerful approach for sample classification and identification of key features influencing classification outcomes, and the integration of multiple machine learning algorithms can further enhance predictive accuracy \u003csup\u003e(\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e)\u003c/sup\u003e. Accordingly, we integrated Random Forest, LASSO regression, and XGBoost algorithms and identified PLA2G2A as a candidate gene by extracting feature importance and intersecting the results across these methods.\u003c/p\u003e \u003cp\u003ePLA2G2A has been reported to exhibit context-dependent and heterogeneous roles across different tumor types. As a phospholipase involved in phospholipid metabolism, PLA2G2A and its major product, arachidonic acid (AA), have been implicated in multiple oncogenic processes \u003csup\u003e(\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e)\u003c/sup\u003e. Previous studies have shown that PLA2G2A is expressed in cancer-associated fibroblasts (CAFs) in breast cancer and pancreatic ductal adenocarcinoma, where it promotes immune cell infiltration \u003csup\u003e(\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e) (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e)\u003c/sup\u003e. In contrast, high PLA2G2A expression has been associated with a significantly reduced risk of recurrence in small cell lung cancer (SCLC) and gastric cancer \u003csup\u003e(\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e) (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e)\u003c/sup\u003e .\u003c/p\u003e \u003cp\u003eIn the present study, by integrating differential expression analysis and survival analysis, we demonstrated that PLA2G2A is significantly upregulated in lymph node metastasis\u0026ndash;positive penile cancer and is positively correlated with lymph node metastasis and poor clinical outcomes. These findings support the identification of PLA2G2A as a key pro-metastatic gene in penile cancer.\u003c/p\u003e \u003cp\u003eSingle-cell transcriptomic profiling revealed that PLA2G2A was specifically expressed in a distinct subpopulation of cancer-associated fibroblasts (CAFs), which we defined as PLA2G2A⁺ CAFs. Although this subset shared certain transcriptional features with progenitor-like CAFs, we observed that it also uniquely expressed CXCL12, suggesting that it may exert its functions through a distinct immunoregulatory network within the penile cancer microenvironment. CAFs are known to recruit myeloid cells via the secretion of chemokines and to induce their immunosuppressive polarization, a finding consistent with the work of Hong Liu et al. \u003csup\u003e(\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e)\u003c/sup\u003e .\u003c/p\u003e \u003cp\u003eCell\u0026ndash;cell communication analysis focusing on PLA2G2A⁺ CAFs demonstrated that this subpopulation exerts broad immune cell recruitment effects through the CXCL12\u0026ndash;CXCR4 axis. More importantly, we found that PLA2G2A⁺ CAFs specifically interact with myeloid cells via the CSF1\u0026ndash;CSF1R pathway, which plays a critical role in macrophage recruitment and the induction of macrophage exhaustion \u003csup\u003e(\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e)\u003c/sup\u003e. To further elucidate the impact of PLA2G2A⁺ CAFs on macrophages, we performed a refined subclustering of myeloid cells and identified monocytes, transitional macrophages, M1 macrophages, M2 macrophages, and proliferating macrophages. Ligand\u0026ndash;receptor interaction analysis further revealed that PLA2G2A⁺ CAFs recruit non\u0026ndash;tumor-associated macrophages (non-TAMs) via CXCL12, while the MDK\u0026ndash;LRP1 and MDK\u0026ndash;NCL signaling axes directly act on both M1 and M2 macrophages. These signaling pathways have been reported to effectively promote macrophage polarization from a pro-inflammatory M1 phenotype toward an anti-inflammatory, tumor-promoting M2 phenotype \u003csup\u003e(\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e) (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e)\u003c/sup\u003e, strongly supporting the notion that PLA2G2A⁺ CAFs are key drivers of immunosuppressive phenotypes within the tumor microenvironment.\u003c/p\u003e \u003cp\u003eWe further conducted pseudotime trajectory analysis of the monocyte\u0026ndash;macrophage population, which clearly delineated the differentiation trajectory from monocytes to macrophages and the polarization process from M1 to M2 macrophages. Cells at the beginning of the trajectory predominantly consisted of monocytes characterized by high expression of FCN1 and VCAN. As pseudotime progressed, these cells transitioned through intermediate states and ultimately differentiated into M2 macrophages with high expression of FOLR2, CD163, and TREM2. Notably, CSF1R expression exhibited an overall increasing trend along pseudotime, indicating progressive macrophage maturation within the tumor. Meanwhile, the receptors for MDK, LRP1 and NCL, maintained relatively high expression levels during the mid-to-late stages of pseudotime (corresponding to the M1-to-M2 transition), which is consistent with our conclusion that PLA2G2A⁺ CAFs drive macrophage polarization via the MDK\u0026ndash;LRP1/NCL axis.\u003c/p\u003e \u003cp\u003eIn addition, cell\u0026ndash;cell communication analysis revealed VEGFA\u0026ndash;VEGFR1 signaling interactions between PLA2G2A⁺ CAFs and endothelial cells. The VEGFA\u0026ndash;VEGFR1 axis is a classical angiogenic pathway that has been well documented to increase vascular permeability, promote tumor angiogenesis, and facilitate tumor cell intravasation. Activation of this pathway has also been closely associated with lymph node metastasis in multiple malignancies \u003csup\u003e(\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e)\u003c/sup\u003e. However, in the specific context of penile cancer, whether PLA2G2A⁺ CAFs directly remodel vascular or lymphatic networks through enhanced VEGFA secretion to mediate lymph node metastasis, as well as the precise downstream molecular mechanisms involved, remain incompletely understood and are rarely reported in the literature. Although the present study did not include in-depth molecular validation of this pathway, this bioinformatic finding suggests that PLA2G2A⁺ CAFs may possess vascular remodeling potential in addition to their immunomodulatory functions. These observations provide important clues and novel directions for future investigations into the complex mechanisms by which PLA2G2A⁺ CAFs cooperate with the vascular system to promote penile cancer metastasis.\u003c/p\u003e \u003cp\u003eAlthough this study identified novel biomarkers for lymph node metastasis in penile cancer, several limitations remain. First, regarding data cohort and model validation, the relatively low incidence of penile cancer, limited sample size, and homogeneous data type inevitably introduce selection bias. More importantly, while we combined multiple machine learning algorithms to improve the accuracy of key feature (PLA2G2A) screening, the lack of large-scale penile cancer cohorts (e.g., large transcriptome or single-cell datasets) with complete follow-up information on lymph node metastasis in public databases prevented us from validating our predictions on independent, multi-center external datasets. Second, in terms of mechanism exploration and functional validation, our conclusions are primarily based on transcriptome-level data mining and algorithmic predictions. Although intercellular communication and pseudo-temporal analysis strongly suggest that PLA2G2A⁺ CAFs may drive macrophage M2 polarization through the MDK-LRP1/NCL axis, these findings mainly reflect associations in gene expression rather than direct causal relationships. How PLA2G2A, as a lipid metabolism enzyme, regulates the secretion of its downstream chemokines remains unclear.\u003c/p\u003e \u003cp\u003eTo overcome these limitations, our future research will proceed in two directions: First, we plan to systematically collect standardized clinical, pathological, imaging, and biological data from newly diagnosed penile cancer patients. Based on this expanded domestic cohort, we will conduct more rigorous clinical validation of the correlation between PLA2G2A and its associated microenvironmental biomarkers and lymph node metastasis to compensate for the lack of external independent datasets. Second, we will dedicate ourselves to conducting in-depth molecular biology experiments using in vitro cell co-culture systems and in vivo animal lymph node metastasis models to ultimately elucidate the precise biological functions of PLA2G2A⁺ CAFs and their associated signaling pathways in driving penile cancer lymph node metastasis.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eAA: Arachidonic Acid\u003c/p\u003e\n\u003cp\u003eAUC: Area Under the Curve\u003c/p\u003e\n\u003cp\u003eBMI: Body Mass Index\u003c/p\u003e\n\u003cp\u003eCAFs: Cancer-Associated Fibroblasts\u003c/p\u003e\n\u003cp\u003eCI: Confidence Interval\u003c/p\u003e\n\u003cp\u003eDFS: Disease-Free Survival\u003c/p\u003e\n\u003cp\u003eEAU: European Association of Urology\u003c/p\u003e\n\u003cp\u003eECM: Extracellular Matrix\u003c/p\u003e\n\u003cp\u003eFFPE: Formalin-Fixed Paraffin-Embedded\u003c/p\u003e\n\u003cp\u003eGEO: Gene Expression Omnibus\u003c/p\u003e\n\u003cp\u003eGO: Gene Ontology\u003c/p\u003e\n\u003cp\u003eHDL: High-Density Lipoprotein\u003c/p\u003e\n\u003cp\u003eHIV: Human Immunodeficiency Virus\u003c/p\u003e\n\u003cp\u003eHPV: Human Papillomavirus\u003c/p\u003e\n\u003cp\u003eHR: Hazard Ratio\u003c/p\u003e\n\u003cp\u003eiCAFs: Inflammatory CAFs\u003c/p\u003e\n\u003cp\u003eIHC: Immunohistochemistry\u003c/p\u003e\n\u003cp\u003eLASSO: Least Absolute Shrinkage and Selection Operator\u003c/p\u003e\n\u003cp\u003eLDL: Low-Density Lipoprotein\u003c/p\u003e\n\u003cp\u003eLNM: Lymph Node Metastasis\u003c/p\u003e\n\u003cp\u003elogFC: Log Fold Change\u003c/p\u003e\n\u003cp\u003emyCAFs: Myofibroblastic CAFs\u003c/p\u003e\n\u003cp\u003eNCCN: National Comprehensive Cancer Network\u003c/p\u003e\n\u003cp\u003eNK cells: Natural Killer cells\u003c/p\u003e\n\u003cp\u003eOS: Overall Survival\u003c/p\u003e\n\u003cp\u003ePC / PeCa: Penile Cancer\u003c/p\u003e\n\u003cp\u003ePCA: Principal Component Analysis\u003c/p\u003e\n\u003cp\u003epHB: Proportions of Hemoglobin genes\u003c/p\u003e\n\u003cp\u003ePLA2G2A: Phospholipase A2 Group IIA\u003c/p\u003e\n\u003cp\u003epMT: Proportions of Mitochondrial genes\u003c/p\u003e\n\u003cp\u003epRP: Proportions of Ribosomal genes\u003c/p\u003e\n\u003cp\u003ePSCC: Penile Squamous Cell Carcinoma\u003c/p\u003e\n\u003cp\u003eROC: Receiver Operating Characteristic\u003c/p\u003e\n\u003cp\u003eSCLC: Small Cell Lung Cancer\u003c/p\u003e\n\u003cp\u003eSM cells: Smooth Muscle cells\u003c/p\u003e\n\u003cp\u003eTAMs: Tumor-Associated Macrophages\u003c/p\u003e\n\u003cp\u003eTCGA: The Cancer Genome Atlas\u003c/p\u003e\n\u003cp\u003eTME: Tumor Microenvironment\u003c/p\u003e\n\u003cp\u003eXGBoost: Extreme Gradient Boosting\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe thank specimen donors of patients of Yunnan Cancer Hospital and research groups that make project data publicly available.\u0026nbsp;We would also like to express our sincere gratitude to the team of Maarten Albersen for providing the rare single-cell data on penile cancer.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe scRNA-seq dataset obtained from Maarten Albersen's group at University Hospitals Leuven, DOI 10.1016/j.eururo.2024.03.038.\u0026nbsp;The data comes from single-cell data of penile cancer published in the European Genome-Phenome Archive (EGA),which is hosted by the EBI and the CRG, under accession number EGAS50000000217.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll authors read and approved the final version of the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDisclosure\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was supported by Scientific Research Fund of Yunnan Provincial Department of Education (no. 2026J0278) and the HengruiResearch Fund of Kunming Medical University (YQHR2025-M18).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of interest disclosure\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors have no relevant conflicts to declare.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthical statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eInformed Consent: All patients signed informed consent forms. Registry and the Registration No. of the study/trial: This study was approved by the Medical Ethics Committee of Yunnan Cancer Hospital (Approval No. (KYLX2022051)).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors contribution\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eChen Hu\u003c/strong\u003e: study design, methodology, writing. \u003cstrong\u003eWeijia Chen\u003c/strong\u003e: data analysis, validation. \u003cstrong\u003eFan Li\u003c/strong\u003e: methodology, validation. \u003cstrong\u003eMinghong Feng\u003c/strong\u003e: formal analysis. \u003cstrong\u003eJianglai Wang\u003c/strong\u003e: literature search. \u003cstrong\u003eZijian Tan:\u003c/strong\u003e investigation. \u003cstrong\u003eBai yu:\u0026nbsp;\u003c/strong\u003eproject supervision, funding acquisition. \u003cstrong\u003eHong Shi\u003c/strong\u003e: project supervision, funding acquisition.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eDouglawi AM, Timothy A. Penile cancer epidemiology and risk factors: a contemporary review. Current opinion in urology. 2019;29(2).\u003c/li\u003e\n\u003cli\u003eOlesen TB, Sand FL, Rasmussen CL, Albieri V, Toft BG, Norrild B, et al. Prevalence of human papillomavirus DNA and p16INK4a in penile cancer and penile intraepithelial neoplasia: a systematic review and meta-analysis. The Lancet Oncology. 2018.\u003c/li\u003e\n\u003cli\u003eWhite B, Dreger D, White BT, Gaul JM. Penile cancer. Nature reviews Disease primers. 2021;7 1:12.\u003c/li\u003e\n\u003cli\u003eKieffer JM, Djajadiningrat RS, Van Muilekom EAM, Graafland NM, Horenblas S, Aaronson NK. Quality of Life for Patients Treated for Penile Cancer. Journal of Urology. 2014;192(4):1105-10.\u003c/li\u003e\n\u003cli\u003eDjajadiningrat RS, Graafland NM, Van Werkhoven E, Meinhardt W, Bex A, Van dP, Henk G., et al. Contemporary Management of Regional Nodes in Penile Cancer\u0026mdash;Improvement of Survival? Journal of Urology. 2014;191(1):68-73.\u003c/li\u003e\n\u003cli\u003ePagliaro LC, Crook J. Multimodality therapy in penile cancer: when and which treatments? World Journal of Urology. 2009;27(2):221-5.\u003c/li\u003e\n\u003cli\u003eWoldu SLC, BoHutchinson, Ryan C.Krabbe, Laura-MariaSingla, NirmishPassoni, Niccolo M.Clinton, Timothy N.Raj, Ganesh V.Miller, David S.Sagalowsky, Arthur I.Lotan, YairXie, YangMargulis, VitalyBagrodia, Aditya. Usage and survival implications of surgical staging of inguinal lymph nodes in intermediate- to high-risk, clinical localized penile cancer: A propensity-score matched analysis. Urologic oncology. 2018;36(4).\u003c/li\u003e\n\u003cli\u003eJeanne-Julien A, Bouchot O, De Vergie S, Branchereau J, Perrouin-Verbe M-A, Rigaud J. Morbidity and risk factors for complications of inguinal lymph node dissection in penile cancer. World Journal of Urology. 2023;41(1):109-18.\u003c/li\u003e\n\u003cli\u003eKroon B, Horenblas S, Lont A, Tanis P, Gallee M, Nieweg O. Patients with penile carcinoma benefit from immediate resection of clinically occult lymph node metastases. Journal of Urology. 2005;173(3):816-9.\u003c/li\u003e\n\u003cli\u003eElst L, Philips G, Vandermaesen K, Bassez A, Lodi F, Vreeburg MT, et al. Single-cell atlas of penile cancer reveals TP53 mutations as a driver of an aggressive phenotype, irrespective of human papillomavirus status, and provides clues for treatment personalization. European Urology. 2024;86(2):114-27.\u003c/li\u003e\n\u003cli\u003eZhang J, Lu T, Lu S, Ma S, Han D, Zhang K, et al. Single-cell analysis of multiple cancer types reveals differences in endothelial cells between tumors and normal tissues. Computational and structural biotechnology journal. 2023;21:665-76.\u003c/li\u003e\n\u003cli\u003eZhang Q, Liu Y, Wang X, Zhang C, Hou M, Liu Y. Integration of single-cell RNA sequencing and bulk RNA transcriptome sequencing reveals a heterogeneous immune landscape and pivotal cell subpopulations associated with colorectal cancer prognosis. Frontiers in Immunology. 2023.\u003c/li\u003e\n\u003cli\u003eVasanthakumar GU, Riadhwsein R, Priyanka GB, Aishwarya B, Kaliappan S, editors. Identification of Lung Cancer Biomarkers from Single-Cell RNA Sequencing Data using Random Forest with Extreme Gradient Boosting. 2025 4th International Conference on Distributed Computing and Electrical Circuits and Electronics (ICDCECE).\u003c/li\u003e\n\u003cli\u003eHidalgo I, Sorolla MA, Sorolla A, Salud A, Parisi E. Secreted Phospholipases A2: Drivers of Inflammation and Cancer. International Journal of Molecular Sciences. 2024;25(22).\u003c/li\u003e\n\u003cli\u003eLiu T, Liu C, Yan M, Zhang L, Zhang J, Xiao M, et al. Single cell profiling of primary and paired metastatic lymph node tumors in breast cancer patients. Nature Communications.\u003c/li\u003e\n\u003cli\u003eWang Y, Liang Y, Xu H, Zhang X, Mao T, Cui J, et al. Single-cell analysis of pancreatic ductal adenocarcinoma identifies a novel fibroblast subtype associated with poor prognosis but better immunotherapy response. Cell Discovery. 2021;7.\u003c/li\u003e\n\u003cli\u003eZhu H, Shi H, Lu J, Zhu K, Yang L, Guo L, et al. Proteomic profiling reveals the significance of lipid metabolism in small cell lung cancer recurrence and metastasis. Journal of Translational Medicine. 2024;22(1):1117.\u003c/li\u003e\n\u003cli\u003eXing XF, Li H, Zhong XY, Zhang LH, Wang XH, Liu YQ, et al. Phospholipase A2 group IIA expression correlates with prolonged survival in gastric cancer. Histopathology. 2011;59(2):198-206.\u003c/li\u003e\n\u003cli\u003eLiu T, Wang Z, Xue X, Wang Z, Zhang Y, Mi Z, et al. Single-cell transcriptomics analysis of bullous pemphigoid unveils immune-stromal crosstalk in type 2 inflammatory disease. Nature Communications. 2024;15(1):5949.\u003c/li\u003e\n\u003cli\u003eWang Q, Wang J, Xu K, Luo Z. Targeting the CSF1/CSF1R signaling pathway: an innovative strategy for ultrasound combined with macrophage exhaustion in pancreatic cancer therapy. Frontiers in Immunology. 2024;15:1481247.\u003c/li\u003e\n\u003cli\u003eZhang Y, Zuo C, Liu L, Hu Y, Yang B, Qiu S, et al. Single-cell RNA-sequencing atlas reveals an MDK-dependent immunosuppressive environment in ErbB pathway-mutated gallbladder cancer. Journal of hepatology. 2021;75(5):1128-41.\u003c/li\u003e\n\u003cli\u003eLiu T, Ding Q, Gou J, Lu C, Lu X, Chen J, et al. Single-cell and spatial transcriptomics integration reveals FAM49B promotes tumor-associated macrophages polarization in colorectal cancer via the MK pathway. Frontiers in Immunology. 2025;16:1682637.\u003c/li\u003e\n\u003cli\u003eLiu W, Xu J, Wang M, Wang Q, Bi Y, Han M. Tumor-derived vascular endothelial growth factor (VEGF)-a facilitates tumor metastasis through the VEGF-VEGFR1 signaling pathway. International journal of oncology. 2011;39(5):1213-20.\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003eTables 1 and 2 are available in the Supplementary Files section.\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Penile Cancer, PLA2G2A, Lymph Node Metastasis, Cancer-Associated Fibroblasts, Biomarker","lastPublishedDoi":"10.21203/rs.3.rs-8936508/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8936508/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eMetastasis to lymph nodes in penile squamous cell carcinoma (PSCC) significantly impacts patient survival rates, and the challenge of accurately identifying patients who would genuinely benefit from inguinal lymph node dissection prior to surgery remains a pressing and unresolved issue in clinical practice. This research utilized single-cell RNA sequencing, bulk RNA sequencing, and various machine learning techniques (including LASSO, Random Forest, and XGBoost) to systematically identify PLA2G2A as a crucial molecular marker for lymph node metastasis in PSCC. In a separate clinical cohort of 69 patients, immunohistochemical studies demonstrated that PLA2G2A was markedly elevated in those with metastasis, and its high levels correlated with unfavorable clinical outcomes. Additionally, we pinpointed PLA2G2A expression to a distinct subpopulation of fibroblasts. Mechanistic investigations indicated that PLA2G2A-positive cancer-associated fibroblasts (CAFs) attract CXCR4-positive monocytes into the tumor microenvironment via CXCL12 secretion, subsequently driving their transformation from an M1-like to an M2-like immunosuppressive phenotype through the MDK\u0026ndash;NCL/LRP1 signaling pathway, thus fostering a tumor microenvironment conducive to metastasis. Overall, this study clarifies the significant regulatory function of PLA2G2A-positive CAFs in PSCC metastasis and positions PLA2G2A as a valuable biomarker for evaluating the risk of lymph node metastasis before surgery.\u003c/p\u003e","manuscriptTitle":"PLA2G2A⁺ CAFs as a Key Driver of Lymph Node Metastasis in Penile Cancer and Its Underlying Mechanisms","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-12 07:38:39","doi":"10.21203/rs.3.rs-8936508/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"22939b29-3fa7-4d0a-9351-e45053b9a8de","owner":[],"postedDate":"April 12th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":65774828,"name":"Health sciences/Biomarkers"},{"id":65774829,"name":"Biological sciences/Cancer"},{"id":65774830,"name":"Health sciences/Oncology"}],"tags":[],"updatedAt":"2026-04-15T10:33:24+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-12 07:38:39","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8936508","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8936508","identity":"rs-8936508","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}