Abstract Objective The ropporin-1 (ROPN1) gene, initially linked to sperm motility, is differentially expressed in triple negative breast cancer (TNBC), suggesting a role in tumor progression and therapy resistance. To characterize ROPN1 expression in breast cancer and evaluate its association with clinicopathological features, survival, and treatment response as a translational biomarker. Materials and Methods Data from The Cancer Genome Atlas (1,087 patients), Sweden Cancerome Analysis Network-Breast (3,273 patients), and geodatabases were analyzed. ROPN1 transcriptional levels were assessed in relation to clinical variables and survival. Chemotherapy agents and epigenetic modulators were tested in cell lines to evaluate ROPN1 regulation. Results Transcriptional overexpression of ROPN1 was significantly enriched in TNBC/basal-like tumors (p<0.0001) and correlated with reduced overall survival, particularly in basal cases [hazard ratio (HR) = 1.85; 95% confidence interval (CI): 1.02–3.33; p = 0.041]. Patients treated with chemotherapy and exhibiting high ROPN1 levels had unfavorable prognosis, with an even poorer profile in untreated cohorts (HR = 4.55; 95% CI: 1.33–14.29; p = 0.01). Hypomethylation at cg00101712 (HR = 0.59; p = 0.016) and cg09298623 (HR = 0.49; p = 0.0014) CpG sites were associated with worse survival at 5 years follow-up, underscoring epigenetic regulation of this pathway as a key driver of poor outcomes. Furthermore, in vitro treatment with cisplatin, doxorubicin, and paclitaxel resulted in variable responses, with a significant reduction of ROPN1 in HCC70 and HS578T cell lines, while BT549 and MDA-MB-231 cell lines showed notable increases. Conclusion ROPN1 overexpression in TNBC/basal-like tumors suggests a role as a prognostic biomarker and predictor of post-chemotherapy resistance. Investigation of ROPN1 expression in breast tumors may lead to alternative strategies targeting pro-metastatic pathways and improve precision treatment for aggressive breast cancer. Keywords: ROPN1, triple negative breast cancer, basal-like, aggressive breast cancer __________________________________________________________________ Key Points • The ropporin-1 gene (ROPN1) overexpression is linked to worse overall survival, especially in triple negative or basal-like breast cancer. • High ROPN1 levels predict poor prognosis in both chemotherapy-treated and untreated patients. • ROPN1 expression inversely correlates with DNA methylation and it is known that hypomethylation is associated with adverse outcomes. • In vitro, cisplatin, doxorubicin, and paclitaxel variably modulated ROPN1 expression in different cancer cell lines with increased expression levels in some cell lines, suggesting therapy resistance. • Treatment with 5-aza-2′-deoxycytidine or trichostatin-A led to increased ROPN1 expression. Introduction Breast cancer remains the most commonly diagnosed malignancy in women, comprising 24.5% of cases with over 2.3 million new diagnoses annually, and is the leading cause of female cancer death (685,000 deaths in 2020) ([31]1). The incidence of breast cancer is higher in developed countries, reflecting both lifestyle factors and advanced screening, whereas delayed diagnosis and limited therapy access drive elevated mortality in low- and middle-income regions ([32]2, [33]3). By 2040, cases are projected to exceed 3 million, with over 1 million deaths ([34]4). From a molecular perspective, breast cancer is a heterogeneous disease classified into distinct subtypes based on gene expression profiles and biomarker presence. The main molecular subtypes include luminal A and B tumors, which express estrogen receptor (ER) and progesterone receptor (PR) in various combinations; human epidermal growth factor receptor 2 (HER2)-enriched tumors, which are characterized by the overexpression of HER2; and triple-negative breast cancer (TNBC), which lacks ER/PR/HER2 expression ([35]5). Molecular classification has been shown to be critical for therapeutic guidance and prognosis since HER2-positive and TNBC tumors typically exhibit more aggressive behavior and differential treatment responses ([36]6). Owing to the heterogeneity of TNBC, Lehmann et al. ([37]7) proposed a TNBC sub-classification system comprising six subtypes: basal-like 1/2, mesenchymal/mesenchymal-like, luminal androgen receptor (LAR), and an immunomodulatory group, each displaying unique molecular profiles with variability in prognosis/treatment sensitivity; some show increased chemotherapy responsiveness, whereas others are correlated with increased relapse risk. Another widely used classification, prediction analysis of microarray 50 (PAM50), categorizes breast cancers into luminal A/B, HER2-enriched, basal-like or normal-like subtypes, using a 50-gene transcriptional panel. This framework is clinically relevant because it provides insights into tumor behavior and aids in therapeutic decision-making ([38]8). Although TNBC is frequently considered synonymous with the basal subtype, studies indicate an approximately 80% overlap between these two categories. This high correlation suggests that different classification techniques may lead to distinct interpretations of tumor behavior. However, given their significant molecular and clinical similarities, it is common for findings from TNBC-focused studies to be extrapolated to basal subtypes and vice versa. This widely adopted practice arises because researchers often access distinct datasets, some of which are based on immunohistochemical classification (TNBC) and others on transcriptional profiling (basal). Thus, extrapolation serves as a pragmatic tool to broaden the applicability of results despite limitations in standardized classification methods ([39]9). The TNBC/basal subtype represents the most challenging entity in oncological management and is characterized by complex molecular heterogeneity, the absence of specific therapeutic targets, high aggressiveness, elevated recurrence rates and reduced five-year overall survival (OS). Despite these therapeutic hurdles, recent advances in targeted immunotherapy with poly (ADP-ribose) polymerase (PARP) inhibitors and combination approaches have emerged as promising strategies ([40]10). In this context, bioprospecting through “omics” data analysis has become an important auxiliary strategy, contributing to novel diagnostic/prognostic biomarker identification, mapping underexplored oncogenic pathways and uncovering potential molecular targets, thereby offering innovative perspectives for understanding this tumor subtype ([41]11). The ropporin-1 (ROPN1) gene was initially identified as a regulator of sperm motility and was first described in human and murine testicular tissues. Current annotations indicate that it encodes a protein predominantly expressed in male reproductive tissues, with functional roles linked to sperm flagellum axoneme formation ([42]12). Subsequent studies revealed ROPN1 expression in diverse reproductive tissues at relatively low levels ([43]13, [44]14). In the context of breast cancer, transcriptomic data suggest differential ROPN1 expression patterns between normal and tumor tissues, with overexpression associated with aggressive malignancies, particularly TNBC subtypes. While the exact mechanistic contribution of ROPN1 to tumorigenesis remains incompletely understood, its established roles in cellular motility pathways may facilitate invasive/metastatic processes ([45]13). Materials and Methods Identification of ROPN1 [46]GSE76275 was interrogated via GEO2R ([47]15, [48]16) to identify DEGs between TNBC and non-TNBC using Benjamini-Hochberg-adjusted p<0.05 and |log[2]FC|≥1.5. ROPN1, among the most dysregulated, was selected for focused TNBC expression and functional analyses. The Cancer Genome Atlas (TCGA) Data Analysis TCGA Firehose Legacy data (n = 1108) were retrieved via cBioPortal, excluding 12 male patients, five without age and four without ROPN1 expression data, yielding 1087 female cases. ROPN1 levels were merged with clinicopathological data and PAM50 calls from Xena by barcode ([49]17). Cases were dichotomized at the median into low (≤median) and high (>median) groups. Categorical associations used χ²; continuous data were log-transformed to z-scores (RNA-Seq V2 RSEM), tested for normality (Shapiro-Wilk) and analyzed by Student’s t/ANOVA or Mann-Whitney/Kruskal-Wallis. HM450 β-values at the ROPN1 locus were compared across PAM50 subtypes and correlated with expression via Spearman’s test. Study Methodology: Sweden Cancerome Analysis Network-Breast (SCAN-B) Analysis RNA-seq from SCAN-B [[50]GSE96058; n = 3,678; n = 3,273 tumors after excluding replicates/NAs; median follow-up 52 months ([51]18)] were processed in R (gtsummary) ([52]19). ROPN1 expression was again dichotomized at the median; Wilcoxon rank-sum and χ² tests assessed clinicopathological associations. Optimal cut-offs were derived via survminer residual-minimization ([53]20) to generate Kaplan-Meier curves and Cox models (survival package; p<0.05), with analyses stratified by basal subtype and chemotherapy status. GEO Database Analysis Five GEO datasets ([54]GSE76275, [55]GSE21653, [56]GSE32646, [57]GSE18864, [58]GSE43358) were retrieved and analyzed via GEO2R to compare ROPN1 probe-specific expression (224191_x_at, 231535_x_at, 233203_at) between groups. Post-analysis outputs were exported and plotted in GraphPad Prism 9 boxplots with consistent scaling and outlier thresholds to visualize cohort-wise expression differences. Functional Enrichment Analysis and Protein-Protein Interaction Network To elucidate ROPN1’s molecular interactions, we performed comprehensive correlation analyses via cBioPortal followed by gene selection for subsequent protein-protein interaction (PPI) network modeling, using STRING-db (https://www.string-db.org/). The selected genes were subjected to rigorous PPI network construction and pathway enrichment analysis with a stringent interaction confidence threshold (≥0.4) to ensure biological relevance ([59]21). Using STRING-db, potential PPI interaction networks were mapped followed by visualization and computational refinement through Cytoscape 3.10.1 (www.cytoscape.org/). MethSurv MethSurv is an R Shiny web portal that uses TCGA CpG beta values (0–1) to perform univariate and multivariate survival analyses with built-in visualization and clustering, with no coding or extra software required ([60]22). In the present study, MethSurv was used to assess survival outcomes using the invasive breast cancer dataset from TCGA, incorporating ROPN1 as a focal point of our analyses. Cell Line Culture and Treatment MCF10A (ATCC CRL-10317) cells were maintained in 1:1 DMEM/F12 supplemented with 5% horse serum, 0.5 µg/mL hydrocortisone, 10 µg/mL insulin, 20 ng/mL EGF and 100 ng/mL cholera toxin. Hs578T (ATCC HTB-126), MDA-MB-231 (ATCC HTB-26), SK-BR-3 (ATCC HTB-30), HCC70 (ATCC CRL-2315), MCF7 (ATCC HTB-22), BT-474 (ATCC HTB-20) and BT-549 (ATCC HTB-122) were cultured in DMEM or RPMI 1640 with 10% FBS and 1% penicillin-streptomycin. All lines were incubated at 37 °C, 5% CO[2], with medium renewed every 48 h until 50–60% confluence. Mycoplasma testing was performed before and after experiments; subcultures used 0.25% trypsin-EDTA. Concentrations of 5-aza-2′-deoxycytidine (5-aza) and trichostatin A (TSA) were set by Alamar Blue assays to avoid morphological or growth alterations. Cisplatin, doxorubicin and paclitaxel were applied at ½ IC[50] [CancerRxGene ([61]23)]; SK-BR-3 dosing was performed as described by Hai et al. A single 6 Gy fraction was delivered via an RS2000 irradiator with Gafchromic dosimetry, followed by 48 h recovery and RNA isolation (SV Total RNA, Promega). cDNA was synthesized from 2 µg RNA (High-Capacity Kit, Thermo Fisher) and quantitative polymerase chain reaction (qPCR) performed (SYBR Green, Applied Biosystems 7500) using primers for ROPN1 ([62]NM_001394219.1; F: 5′-CCAAAGCCGCCATTAGGGT-3′, R: 5′-GGCTGCCCACTGGATGAG-3′) and GAPDH ([63]NR_152150.2; F: 5′-GACTGTGGTCATGAGTCCTCCC-3′, R: 5′-CAAGATCATCAGCAATGCCTCC-3′). Relative expression was normalized to GAPDH and calculated by ΔΔC[t] in triplicate. Statistical Analysis Statistical analysis was performed using specialized software. This included SPSS, version 25.0 (IBM Inc., Armonk, NY, USA) and GraphPad, version 7 (California, USA). Data normality was assessed by Shapiro-Wilk test. Categorical variables were compared using χ² test; continuous variables employed Student’s t-test or ANOVA for normally distributed data and Mann-Whitney or Kruskal-Wallis tests otherwise. Associations were evaluated by Spearman’s (non-parametric) rank correlation. Survival outcomes were estimated via Kaplan-Meier curves with log-rank testing and multivariate hazard ratios (HRs) calculated by Cox proportional hazards regression. All tests were two-tailed with significance defined as p<0.05. Results Differences Observed in TCGA Data It was observed that high ROPN1 expression was significantly associated with key clinical and pathological characteristics. Patients in the high ROPN1 subgroup were more frequently premenopausal (p = 0.0027) and exhibited a predominance of hormone receptor-negative tumors (ER-/PR-, p<0.0001) and HER2-negative status (p<0.0001). Furthermore, high ROPN1 expression levels were strongly correlated with basal-like and normal-like subtypes based on the PAM50 classification (p<0.0001). Among patients with TNBC, the majority exhibited high ROPN1 expression (p<0.0001). Differences in histological type were significant (p = 0.0013), whereas TNM staging did not show significant variation between groups (p = 0.2276) ([64]Table 1). Table 1. Clinicopathological characteristics of patients with breast cancer derived from the TCGA database and their associations with ROPN1 expression levels. Variables Low High p n - n - Age ≤50 132 24.30 201 37.00 <0.0001 >50 412 75.70 342 63.00 - Menopause status Pre 101 20.60 129 26.70 0.0027 Peri 13 2.60 27 5.60 - Post 377 76.80 327 67.70 - Cancer type detailed IDC 418 77.00 383 70.50 0.0013 ILC 80 14.70 126 23.20 - Other 45 8.30 34 6.30 - TNM stage Stage 1 80 15.00 100 18.80 0.2276 Stage 2 309 58.10 307 57.80 - Stage 3 131 24.60 117 22.00 - Stage 4 12 2.30 7 1.30 - ER status by IHC Negative 44 8.60 194 37.00 <0.0001 Positive 469 91.40 330 63.00 - PR status by IHC Negative 113 22.10 228 43.60 <0.0001 Positive 398 77.90 295 56.40 - HER2 status by IHC Negative 250 70.00 308 85.10 <0.0001 Positive 107 30.00 54 14.90 - PAM50 classification Basal 5 1.10 135 35.20 <0.0001 HER2 45 9.90 22 5.70 - Luminal A 243 53.60 173 45.20 - Luminal B 158 34.90 31 8.10 - Normal-like 2 0.40 22 5.70 - TNBC status nTNBC 496 98.2 356 76.9 <0.0001 TNBC 9 1.8 107 23.1 - [65]Open in a new tab ER: Estrogen receptor; HER2: Human epidermal growth factor receptor 2; IDC: Invasive ductal carcinoma; IHC: Immunohistochemistry; ILC: Invasive lobular carcinoma; PAM50: Prediction analysis of microarray 50 (50-gene panel used for molecular classification); Peri: Perimenopausal; Post: Postmenopausal; PR: Progesterone receptor; Pre: Premenopausal; TNBC: Triple-negative breast cancer; TNM: Tumor, node, metastasis (staging system for tumor size, lymph node involvement, and metastasis); TCGA: The Cancer Genome Atlas The pattern of ROPN1 expression was analyzed in different clinicopathological contexts, as illustrated in the graphs ([66]Figure 1A-F). In each graph, the expression of ROPN1 was compared among subgroups with distinct clinical characteristics and significant variations emerged. ROPN1 mRNA expression was higher in ER-negative, PR-negative, and HER2-positive tumors compared with their opposite counterparts (all p<0.0001; [67]Figure 1A-C). Expression varied by menopausal status (p = 0.0171; [68]Figure 1D) and was elevated in invasive ductal carcinoma versus other histological types (p = 0.0126) ([69]Figure 1E). Basal-like tumors showed the highest ROPN1 levels among molecular subtypes (p<0.0001) ([70]Figure 1F). Inverse correlation between ROPN1 mRNA levels and promoter methylation was observed in unstratified TCGA breast tumors ([71]Figure 2A; r = −0.41; p<0.0001) and was stronger in basal-like tumors ([72]Figure 2B; r = −0.55; p<0.0001). Stratification by clinical subtype highlighted significantly lower methylation in basal-like versus HER2-enriched, luminal A/B and normal-like tumors ([73]Figure 2C, [74]D). Survival analysis via MethSurv showed that hypermethylation at cg00101712 and cg09298623 correlated with improved prognosis [[75]Figure 2E; HR = 0.59; 95% confidence interval (CI): 0.38–0.92; p = 0.016; [76]Figure 2F; HR = 0.49; 95% CI: 0.31–0.78; p = 0.0014]. Figure 1. [77]Figure 1 [78]Open in a new tab ROPN1 expression patterns in different clinical-pathological contexts of breast cancer. The graphs show comparisons of ROPN1 expression among subgroups for (A) estrogen receptor, (B) progesterone receptor, (C) HER2, (D) menopausal status, (E) histological type, and (F) molecular subtypes according to the PAM50 classification. Dunn’s multiple comparisons test was applied, and differences between the basal and HER2, luminal A or luminal B subtypes were significant (p<0.0001), whereas only the difference between the basal and normal-like subtypes was not significant (p = 0.2751) ER: Estrogen receptor; PR: Progesterone receptor; HER2: Human epidermal growth factor receptor type II; Peri: Perimenopausal; Pre: Premenopausal; Post: Postmenopausal; IDC: Invasive ductal carcinoma; ILC: Invasive lobular carcinoma; B: Basal; H: HER2; LA: Luminal A; LB: Luminal B; N: Normal-like Figure 2. [79]Figure 2 [80]Open in a new tab Methylation profile of ROPN1 in breast cancer patients. (A) Correlation between ROPN1 mRNA levels and methylation in a population of breast cancer patients. (B) Specific correlation for the basal subtype of breast cancer, where the negative correlation is more pronounced. (C) Methylation pattern of ROPN1 according to the PAM50 classification. (D) Comparison of ROPN1 methylation in the clinical profiles of basal and non-basal breast cancer patients. (E) OS analysis based on the methylation pattern of the ROPN1 body opening site cg00101712. (F) OS analysis based on the methylation profile of ROPN1 in TSS1500 N-shore cg09298623. (G) The PPI network was constructed using the STRING database, employing the top 40 genes correlated with ROPN1 from the TCGA Firehose Legacy database. Red represents genes positively correlated with ROPN1, whereas green indicates genes negatively correlated with ROPN1. (H) Pathway enrichment analysis B: Basal; H: HER2; LA: Luminal A; LB: Luminal B; N: Normal-like; PPI: Protein-protein interaction; PAM50: Prediction analysis of microarray 50; HER2: Human epidermal growth factor receptor 2 Global correlation via cBioPortal identified 14,462 ROPN1-associated genes; the top 20 positive and 20 negative correlates were input into STRING to generate a PPI network ([81]Figure 2G). Positively linked nodes included ROPN1B, SOX10, FABP7, SOSTDC1, SFRP1, BCL11A, FOXC1 and MIA; negatively linked nodes comprised BCAS1, GATA3, AR, ARMT1, FOXA1, XBP1, WWP1, ESR1 and TMBIM6. STRING enrichment highlighted glandular morphogenesis and hormonal response pathways, prostate gland epithelium and glandular acinus development; branched and epithelial tube morphogenesis; and cellular response to estrogenic stimuli, underscoring the interplay of gland architecture and hormone signaling in breast cancer progression ([82]Figure 2H). Prognostic Insights from the SCAN-B Study on ROPN1 Expression in Breast Cancer SCAN-B RNA-seq data recapitulated TCGA associations: High (n = 1,636) versus low (n = 1,637) ROPN1 expression groups differed in age ≤55 y (36%; p<0.001), tumor size ≤17 cm (56%; p<0.001), ER+ (89% versus 96% and PR+ 84% versus 90%; p<0.001), luminal A enrichment in high expression (55%) and luminal B/HER2 in low expression (p<0.001), endocrine therapy use in high ROPN1 expression (71%; p<0.001) and chemotherapy in low expression (62%; p = 0.031) groups; nodal status was comparable (63% negative; p>0.9) ([83]Supplementary Table 1). ROPN1 mRNA was elevated in ER−, PR− and HER2− tumors (all p<0.0001), and in tumors with high Ki-67 staining (p = 0.0191), as well as in tumors in which endocrine therapy was not used (p<0.0001). Expression was highest in basal and HER2-enriched subtypes, decreasing through luminal A and B and normal-like (p<0.0001), indicating an association with aggressive phenotypes ([84]Supplementary Figure 1). High ROPN1 expression predicted poorer OS in SCAN-B ([85]Figure 3A; HR = 2.17; 95% CI: 1.61–2.94; p<0.0001), with a pronounced effect in basal tumors ([86]Figure 3B; HR = 1.85; 95% CI: 1.02–3.33; p = 0.041). In chemotherapy-treated patients, high ROPN1 remained adverse ([87]Figure 3C; HR = 2.86; 95% CI: 1.28–6.25; p = 0.01), and in untreated patients the mortality risk was even higher ([88]Figure 3D; HR = 4.55; 95% CI: 1.33–14.29; p = 0.01). Figure 3. [89]Figure 3 [90]Open in a new tab Associations between ROPN1 expression and survival in breast cancer patients. (A) Overall survival analysis of all breast cancer patients. (B) Stratified survival of patients with the basal subtype of breast cancer. (C) Impact of ROPN1 expression on the survival of basal breast cancer patients undergoing chemotherapy. (D) Comparative survival analysis of basal breast cancer patients who did not receive chemotherapy. Data were obtained from the [91]GSE96058 study and analyzed via Kaplan-Meier curves. The “Low” and “High” categories refer to patient classification based on ROPN1 gene expression levels, with “Low” indicating expression below the established cut-off value, and “High” indicating expression above this threshold. Expression pattern of ROPN1 in different cohorts. The probes 224191_x_at (E), 231535_x_at (F), and 233203_at (G) were evaluated using datasets available in the GEO database, identified as [92]GSE76275, [93]GSE21653, [94]GSE32646, [95]GSE18864, and [96]GSE43358, respectively Probes 224191_x_at ([97]Figure 3E), 231535_x_at ([98]Figure 3F) and 233203_at ([99]Figure 3G) showed consistent ROPN1 overexpression in TNBC versus non-triple negative subtypes across [100]GSE76275, [101]GSE21653, [102]GSE32646, [103]GSE18864 and [104]GSE43358, underscoring its association with invasive tumor phenotypes. Transcriptional Modulation of ROPN1 in Breast Cancer Cell Lines After culturing non-tumoral mammary tissue cell lines and other representative malignant phenotypes were cultured, total RNA extraction was performed for reverse transcription-qPCR analysis. When MCF10A was used as a reference, SKBR3 and MCF7 cells exhibited ROPN1 expression levels that were more than 2,000 times greater. The BT474, HS-578T, BT549, and MDA-MB-231 lines expressed increases ranging from 0.5 to approximately 60 times. Notably, among those analyzed, HCC70 demonstrated a ROPN1 expression level over 7,000 times greater than that of MCF10A, highlighting the increased expression of this gene in certain tumor contexts ([105]Figure 4A). Figure 4. [106]Figure 4 [107]Open in a new tab Expression of ROPN1 in breast cancer cell lines and the effects of different treatments. (A) Basal transcription levels of ROPN1 in a panel of mammary cell lines, including distinct tumor subtypes. (B-H) Variation in ROPN1 expression after treatment with cisplatin, doxorubicin, paclitaxel, or radiotherapy. (I) Modulation of ROPN1 expression in response to the epigenetic-acting drugs 5-aza and trichostatin A The cell lines were subsequently treated with chemotherapeutics and radiotherapy to evaluate any transcriptional modifications. MCF7 cells presented a greater fold change than did wild-type cells (untreated), with increases of 23%, 3%, and 26% after exposure to cisplatin, doxorubicin, and paclitaxel, respectively. In contrast, irradiation with 6 Grays resulted in a drastic decrease in ROPN1 ([108]Figure 4B). For the SKBR3 line, there was a decrease of approximately 40% after both cisplatin and paclitaxel treatment, although doxorubicin and radiation did not reduce ROPN1 expression levels in this cell line ([109]Figure 4C). In BT474 cells, cisplatin and doxorubicin induced decreases of 48% and 38%, respectively. Conversely, irradiation led to a 25% increase in the transcript level ([110]Figure 4D). For the HCC70 line, which, among the lines used, was the one that expressed ROPN1 at the highest level, all the treatments induced a decrease in the level of this transcript, which was most evident with cisplatin, with which a 63% reduction was achieved ([111]Figure 4E). Similarly, HS578T cells also exhibited a reduction of approximately 50% in response to cisplatin, doxorubicin, and irradiation and a 30% decrease after paclitaxel treatment ([112]Figure 4F). BT549 cell expression of ROPN1 was were reduced by 21% only after cisplatin treatment and significantly increased by 197% and 293% after doxorubicin and paclitaxel treatment, respectively ([113]Figure 4G). Finally, for MDA-MB231 cells, in which the expression levels were close to those found in the MCF10A reference cells, all the treatments induced a significant increase in ROPN1 ([114]Figure 4H). The HS578T line was also treated with 5-aza and TSA, which resulted in increases in the expression of ROPN1 by 9.5-fold and 4.9-fold, respectively ([115]Figure 4I). Discussion and Conclusion Breast cancer remains a leading cause of female cancer mortality, with TNBC/basal tumors exhibiting high recurrence and chemoresistance. Biomarker discovery is therefore essential for enhanced risk stratification and therapeutic decision-making. ROPN1 emerged as a candidate, showing elevated expression in TNBC/basal cases and consistent association with poor survival. Our data indicate that high ROPN1 expression is consistently associated with poor OS in breast cancer patients, especially those with basal-like subtype, which is known for its high degree of aggressiveness and poor prognosis. Risk analysis showed that patients with high ROPN1 expression who were not treated with chemotherapy had an HR of 4.55 ([116]Figure 3D), indicating a robust association between ROPN1 expression and unfavorable outcomes. Among patients who received chemotherapy, high levels of ROPN1 also correlated with poorer prognosis (HR = 2.53). These findings position ROPN1 as a potent prognostic marker of aggressive breast cancer and a potential predictor of limited chemotherapy response, underscoring the need for personalized treatment strategies. Previous findings by Liu et al. ([117]13) showed that ROPN1 is overexpressed in TNBC, enhancing migration, invasion and metastasis via RhoA activation. Overexpression increased actin stress fibers and contractility, while ROPN1 silencing reduced invasiveness and metastasis in vitro and in vivo. Kortleve et al. ([118]24) validated ROPN1 as a prognostic and therapeutic target in TNBC, showing strong expression in >90% of primary and metastatic tumors, with minimal expression in normal tissues, except testis. In patient-derived and murine models, anti-ROPN1 TCR-T cells effectively eliminated ROPN1+ tumors, outperforming cisplatin and sacituzumab govitecan. ROPN1 expression correlated with metastasis and poor prognosis. These findings align with our results, supporting the translational value of ROPN1 as a biomarker and therapeutic target in TNBC. In our PPI analysis, ROPN1B and SOX10 emerged as key genes positively correlated with ROPN1. ROPN1B, a cytoskeleton-related protein sharing 96% sequence homology with ROPN1, may act synergistically to promote invasiveness. Da Gama Duarte et al. ([119]25) reported a strong correlation between ROPN1 and ROPN1B in melanoma (r = 0.86, p = 8.71×10^-4), associating both with motility, chemoresistance and immune modulation. SOX10, which also correlated with ROPN1, drives mesenchymal traits and drug resistance in TNBC ([120]26, [121]27, [122]28), suggesting that their co-expression may enhance tumor plasticity and therapeutic evasion. The network also included clinically relevant genes, such as ESR1, AR and FOXA1, all downregulated or absent in TNBC/basal tumors. ESR1, a key marker in luminal subtypes, serves as both a prognostic indicator and therapeutic target. AR, though expressed in a subset of TNBCs, lacks the favorable impact seen with ESR1 but may serve as a target in LAR tumors ([123]10). FOXA1, a transcriptional cofactor for both ESR1 and AR, promotes epithelial identity in luminal cancers but is minimally expressed in TNBC, supporting its undifferentiated and aggressive profile ([124]29, [125]30). Chemotherapeutic treatments and radiotherapy have heterogeneous effects on ROPN1 levels, depending on the cell line and the therapeutic agent used. Among the parental lines analyzed, HCC70 exhibited an increase in ROPN1 expression of more than 7,000 times, highlighting the unique behavior of the TNBC subtype, which is highly aggressive and has a poor prognosis ([126]7). This overexpression may be associated with the role of ROPN1 in fundamental biological processes, such as cell motility, which directly contributes to tumor migration and invasion. Indeed, previous studies have demonstrated that ROPN1 plays an active role in tumor progression, especially in TNBC. Wu and collaborators reported that the overexpression of ROPN1 was associated with a significant increase in cell migration and invasion, which is mediated by the activation of RhoA, a GTPase essential for cytoskeletal organization ([127]24). This activation leads to actin reorganization, promoting the formation of cellular protrusions and facilitating tumor spread. Furthermore, in vivo models have shown that ROPN1 not only enhances metastasis but also that its suppression significantly reduces tumor dissemination ([128]24). By analyzing the effects of chemotherapeutic agents on a selection of cell lines, we propose a relevant translational hypothesis. We observed that, under certain conditions, treatments did not reduce or even increased the levels of ROPN1, as noted in BT549 cells after exposure to doxorubicin and paclitaxel. Considering the findings of Wu et al. on the prometastatic role of ROPN1, these results suggest that the persistence or elevation of this transcript following chemotherapy may represent an adaptive mechanism of tumor cells, promoting therapeutic resistance and enhancing their migratory and survival capabilities. This hypothesis becomes even more pertinent when the clinical survival data of patients treated with chemotherapy are reviewed ([129]Figure 3C), where high expression of ROPN1 was associated with a poorer prognosis, even after treatment. The observed relationship between ROPN1 expression and its methylation in breast cancer suggests that epigenetic mechanisms may be involved in the regulation of this gene, particularly in the context of more aggressive subtypes, such as HER2-positive breast cancer and TNBC/basal cancer. This pattern of hypomethylation associated with ROPN1 overexpression indicates that DNA methylation might act as a modulator of gene expression in highly proliferative cancers with increased invasive capacity. To date, only the study by Atanackovic and colleagues has evaluated possible associated epigenetic mechanisms through pharmacological treatment in cell lines derived from acute myeloid leukemia (AML) ([130]31). In their study, treatment with TSA and decitabine or their combination did not result in positive modulation of ROPN1, which is poorly expressed in AML, suggesting that other regulatory mechanisms may occur in these lines, such as post-transcriptional regulation mediated by microRNAs or interactions with inhibitory transcription factors. Of note, our in vitro experiments revealed a significant increase in ROPN1 expression after treatment with 5-aza or TSA, with elevated levels of 9.5- and 4.9-fold, respectively. These results reinforce our in silico findings and suggest a distinct epigenetic role of ROPN1 in breast cancer, especially considering the contrast with the AML model. These data highlight the importance of investigating how the epigenetic deregulation of ROPN1 influences tumor behavior in different biological contexts, especially in more aggressive breast subtypes, and exploring whether hypomethylation and overexpression of this gene have direct functional impacts on tumor progression and therapeutic response. Moreover, we observed that low methylation in the regions cg09298623 and cg00101712 was associated with worse OS in patients with this phenomenon. The increased expression of ROPN1, which is mediated by a reduction in methylation, may represent a phenotypic adaptation of the tumor that facilitates cell invasion and metastasis, which can lead to a worsened clinical outcome for these patients. Further studies are needed to clarify how epigenetic changes drive tumor aggressiveness and to define the functional role of ROPN1 in breast cancer progression and therapy response. While our multicohort analysis offers strong associative evidence, functional validation (e.g., CRISPR/Cas9) is required to confirm causality in chemoresistance. Prospective cohorts and mechanistic studies will be key to validating ROPN1 expression as a predictive biomarker, potentially guiding patient stratification and improving therapeutic outcomes in high-risk cases. Study Limitations This study has limitations, including the absence of validation in patient-derived tumor samples and reliance on public gene expression datasets. In addition, the interactions between ROPN1, tumor subtypes, and therapeutic contexts, especially drug combinations, remain incompletely characterized. Further studies are needed to validate these findings and clarify the role of ROPN1 in breast cancer biology. ROPN1 is markedly overexpressed in hormone receptor–negative and triple-negative/basal-like breast cancers, where it predicts significantly poorer OS. Its prognostic value persists regardless of chemotherapy status, high ROPN1 expression doubles mortality risk in treated patients and quadruples it when treatment is absent, underscoring its utility for risk stratification. An inverse relationship between DNA methylation and ROPN1 expression further links hypomethylation to adverse outcomes. In vitro, ROPN1 expression following chemotherapeutics and radiotherapy was variable in different cell lines, with some agents inducing upregulation suggesting adaptive resistance mechanisms. Collectively, these findings position ROPN1 and its protein product as both a robust prognostic and predictive biomarker and a candidate therapeutic target for high-risk breast cancer subgroups. Ethics Ethics Committee Approval: Not necessary. Informed Consent: Not necessary. Supplementary Materials Supplementary Figure 1 Expression pattern of ROPN1 in patients with breast cancer from the Sweden Cancerome Analysis Network-Breast study. Transcriptional levels of ROPN1 based on (A) estrogen receptor, (B) progesterone receptor, (C) human epidermal growth factor receptor 2, (D) the proliferation marker Ki-67, (E) endocrine treatment, and (F) Prediction analysis of microarray 50 classification [131]EurJBreastHealth-21-4-345-supplementaryfigure-1.png^ (299.2KB, png) Supplementary Table 1. Clinicopathological characteristics of patients with breast cancer derived from the Sweden Cancerome Analysis Network-Breast study and their associations with ROPN1 expression. Variables High - Low - p n = 1.636 % n = 1.637 % Age <0.001 ≤55 527 36 372 25 - >55 955 64 1.115 75 - Tumor size group <0.001 ≤17 cm 813 56 705 48 - >17 cm 651 44 768 52 - Lymph node status >0.9 Negative 906 63 910 63 - Positive 533 37 531 37 - ER status <0.001 No 152 11 62 4 - Yes 1.198 89 1.371 96 - PR status <0.001 No 205 16 142 10 - Yes 1.091 84 1.219 90 - HER2 status 0.009 No 1.261 88 1.229 85 - Yes 164 12 214 15 - Ki-67 status 0.001 No 318 46 250 37 - Yes 372 54 420 63 - NHG <0.001 G1 259 18 190 13 - G2 680 47 711 48 - G3 505 35 565 39 - PAM50 subtype <0.001 Basal 279 19 29 2 - HER2 91 6 204 14 - Luminal A 821 55 680 46 - Luminal B 103 7 563 38 - Normal-like 188 13 11 0.7 - Chemo treated 0.031 No 848 58 911 62 - Yes 622 42 568 38 - Endocrine treated <0.001 No 424 29 228 15 - Yes 1.045 71 1.251 85 - [132]Open in a new tab Pearson’s chi-squared test; ER: Estrogen receptor; PR: Progesterone receptor; HER2: Human epidermal growth factor receptor 2; PAM50: Prediction analysis of microarray 50 (50-gene panel used for molecular classification); NHG: Nottingham histologic grade Acknowledgments