BACKGROUND:C-reactive protein (CRP) is a biomarker of vascular inflammation with prognostic value for future cardiovascular events. This meta-analysis investigates the effect of colchicine, a cheap and widely available anti-inflammatory medication, on CRP levels in the months following myocardial infarction (MI). METHODS:PubMed, EMBASE, and Cochrane were queried from inception to April 2026 to identify randomized controlled trials comparing colchicine to placebo for at least 1 month following MI. The primary outcome was mean CRP level at follow-up. Effect estimates were pooled with random-effects models and reported as mean differences for continuous variables using 95% confidence intervals. RESULTS:Four studies met inclusion criteria comprising 3384 patients (mean age 60.7 years; 78.3% male), including 1669 patients randomized to the colchicine arm, and 1715 to placebo. Median follow-up period was 3 months (range: 1-6 months). Colchicine following MI resulted in a statistically significant decrease in mean CRP level (mg/L) at follow-up versus placebo (MD: -0.69; [-1.21, -0.17], p = 0.009). Heterogeneity of effect size estimates was high (I2 = 97%). CONCLUSION:Daily colchicine following MI decreases CRP at a median follow-up period of 3 months compared to placebo. The correlation between CRP reduction with colchicine and adverse cardiovascular event reduction warrants additional study.
BACKGROUND:Early diagnosis of acute mesenteric ischemia (AMI) remains challenging due to the lack of reliable biomarkers. This study aimed to investigate the time-dependent changes in chitinase-3-like protein 1 (YKL-40) and selected inflammatory markers in an experimental AMI model. METHODS:A randomized controlled experimental study was conducted on 48 Wistar-Albino rats. Mesenteric ischemia was induced by ligation of the superior mesenteric artery. Blood and tissue samples were collected at 1, 3, and 6 hours. Serum YKL-40, interleukin-6 (IL-6), interleukin-10 (IL-10) and lactate levels were measured, and histopathological damage was assessed. RESULTS:IL-6 levels increased significantly as early as 1 hour (p = 0.010) and showed a progressive rise over time (p < 0.001), with a strong positive correlation with histopathological damage (r = 0.805). Lactate levels increased significantly at later time points. YKL-40 levels did not differ in the early phase but were significantly elevated at 6 hours (p < 0.001). IL-10 levels decreased significantly in advanced ischemia (p = 0.005). CONCLUSION:IL-6 showed an early increase and a strong association with histopathological damage, whereas YKL-40 appeared to be associated with later-stage inflammation rather than early ischemic injury. These findings highlight the importance of time-dependent biomarker interpretation and support further evaluation of combined biomarker strategies in AMI.
INTRODUCTION:Endometriosis is a chronic inflammatory disease in which monocytes/macrophages play a key pathogenetic role. The peritoneal fluid (PF) cytokine profile can modulate recruited monocyte functions, determining their contribution to disease progression. AIM:To investigate stage-specific changes in the PF cytokine profile and their impact on functional reprogramming of monocytes in endometriosis. METHODS:The study included 26 patients with endometriosis (stages I-III) and 6 controls. PF cytokine concentrations were measured by multiplex analysis (Bio-Plex). The effect of PF on MerTK, CCR2, PD-1 and PD-L1 expression by healthy donor monocytes was assessed by flow cytometry. RESULTS:PF composition changed dynamically with disease stage: At stage I, RANTES and IL-6 increased; at stage II, IL-12 decreased and IL-1ra increased; and at stage III, IFN-γ and IL-12 decreased. Stage I PF inhibited MerTK expression on monocytes, whereas stages II-III PF upregulated MerTK, PD-L1, PD-1 and CCR2. Pro-inflammatory cytokines negatively correlated with MerTK and PD-1 changes, while Th2 cytokines positively correlated with MerTK expression. Th2 cytokine levels were directly associated with serum CA-125 concentration. CONCLUSION:Stage-specific shifts in the PF cytokine profile initiate the switching of monocytes from a pro-inflammatory to an immunosuppressive phenotype. CLINICAL SIGNIFICANCE:Endometriosis diagnosis remains challenging due to the limited sensitivity of CA-125, particularly in early-stage disease. This study demonstrates that stage-specific shifts in peritoneal fluid cytokine and chemokine profiles - including changes in RANTES, IL-6, IL-12, IFN-γ and IL-1ra - drive measurable reprogramming of monocyte surface markers (MerTK, PD-1, PD-L1 and CCR2). These exploratory findings suggest that profiling peritoneal fluid cytokines together with monocyte phenotypic markers may, in principle, provide additional information relevant to disease staging and to the future identification of immune biomarkers; however, the present cohort is small and single-centre, and these observations require validation in larger, independent cohorts before any diagnostic or clinical application can be proposed. Delineating the stage-dependent transition from a pro-inflammatory to an immunosuppressive monocyte phenotype may also help to inform future research into immunomodulatory approaches in endometriosis.
AIM:This study aimed to evaluate the diagnostic utility and prognostic significance of serum phoenixin-14 (PNX-14) and phoenixin-20 (PNX-20) levels in patients with endometrial cancer (EC). METHODS:A prospective case-control study was conducted, enrolling 52 patients with histologically confirmed EC and 45 healthy controls. Serum PNX-14 and PNX-20 concentrations were measured using enzyme-linked immunosorbent assay. Multivariable logistic regression and receiver operating characteristic (ROC) curve analyses were performed. Associations between phoenixin levels and clinicopathological parameters were evaluated. RESULTS:Serum PNX-14 (80.70 ± 22.45 vs. 61.75 ± 18.73 pg/mL; p < 0.0001) and PNX-20 (888.2 ± 284.9 vs. 646.7 ± 243.8 pg/mL; p < 0.0001) levels were significantly elevated in the EC group. Elevated PNX-14 was significantly associated with distant metastasis (p = 0.0463) and disease recurrence (p = 0.0020). Multivariable logistic regression identified PNX-14 as an independent predictor for EC (OR: 1.07; 95%CI: 1.03-1.12; p = 0.001). The combined PNX-14/PNX-20 model yielded a superior area under the curve of 0.844. CONCLUSIONS:Serum PNX-14 and PNX-20 are significantly elevated in EC and demonstrate diagnostic potential. PNX-14 may serve as a prognostic biomarker for metastasis and recurrence.
BACKGROUND:The role of eosinophils in acute appendicitis (AA) is not well known. The aim of this study is to analyze whether peripheral blood eosinophil concentration can be useful in predicting episodes of complicated AA in both paediatric and adult populations. METHODS:A retrospective cohort study was conducted in patients with a clinical diagnosis of AA confirmed by histopathological findings. Receiver operating characteristic (ROC) curve analysis and multivariate models were performed. RESULTS:A total of 531 cases were included (43.7% children). Complicated AA was present in 31.5% of children and 34.4% of adults. Symptom duration among children was ≥36 hours (aOR 2.24, p = 0.006) and a neutrophil-to-lymphocyte ratio (NLR) ≥4.7 (aOR 3.63, p = 0.006) increased the likelihood of complicated AA. In adults, eosinophil concentration <40 cells/µL (aOR 1.88, p = 0.033), being male (aOR 2.53, p < 0.001), and having an NLR ≥4.7 (aOR 2.98, p = 0.006) were independent factors associated with complicated AA. CONCLUSIONS:Peripheral blood eosinophils appear to serve as a valuable biomarker for identifying cases of complicated AA among the adult population but not among children. Prospective studies will help elucidate this phenomenon.
BACKGROUND:Acute respiratory toxicity is a common presenting emergency in acutely poisoned patients. Unpredictable respiratory deterioration can occur regardless of the on-admission patients' stable state, leading to increased morbidity and mortality. This study aimed to evaluate the respiratory rate-oxygenation (ROX) index, heart rate, acidosis, consciousness, oxygenation and respiratory rate (HACOR) score and C-reactive protein (CRP) as predictors of mechanical ventilation (MV) in poisoned patients. METHODS:This prospective study was conducted on poisoned patients with acute respiratory failure presented to the Poison Control Center-Ain Shams University Hospitals (PCC-ASUHs). Upon admission, all patients had their ROX index, HACOR score and CRP level measured at the time of admission and at 24 hours. RESULTS:Seventy-two patients were enrolled in the study, where forty-one cases were mechanically ventilated. The initial ROX index was significantly lower in the mechanically ventilated group with a cut-off ≤ 18.85 at AUC (0.74). The 24-hour ROX index, HACOR score and CRP level were predictors of mechanical ventilation need at AUC (0.97, 0.94, 0.85, respectively). CONCLUSION:It was concluded from this study that the initial and 24-hour ROX index, 24-hour HACOR and CRP level can be predictors of MV in poisoned patients.
BACKGROUND:Sepsis can induce multiple organ dysfunction, and its complications, such as acute kidney injury, are associated with poor prognoses. Serum pro-enkephalin (pro-ENK), as a stable biomarker, holds potential value for risk stratification and prognostic assessment in sepsis. This review aimed to investigate its value through a meta-analysis. METHODS:Databases (PubMed, Embase, Cochrane Library, Web of Science) were searched from database establishment to June 16, 2025, and literature was screened according to the established PICOS principles. Cohort studies that examined the relationship between pro-ENK levels and poor prognosis were included. Data were extracted, and study quality was appraised via the NOS (Newcastle-Ottawa Scale) assessment tool. This meta-analysis was conducted using Stata 15.0. RESULTS:9 studies were included, with sample sizes ranging from 40 to 956 participants per study. In sepsis or shock patients, pro-ENK levels were higher in the AKI group (SMD: 1.82, 95% CI: 1.26-2.39); pro-ENK levels were higher in the non-survivor group among patients with sepsis or shock (SMD: 0.85, 95% CI: 0.60-1.10). AKI occurrence was associated with penKid levels (OR: 2.50, 95% CI: 2.02-2.99). CONCLUSION:Current evidence identifies pro-ENK as an auxiliary biomarker for sepsis-related AKI and its prognosis and provides valuable insights for further clinical research.
BACKGROUND:Acute mesenteric ischemia (AMI) is difficult to recognize and is often diagnosed late. Currently there are no established biomarkers that can assist in diagnosis or management. The present study was a systematic review of animal data, intended to identify potentially time-sensitive biomarkers for human AMI. METHODS:A systematic literature search was conducted using PubMed, Web of Science, and Scopus up to 10/02/25. Studies assessing potential biomarkers of AMI in mammalian models, with timed responses from the onset of experimental ischemia were selected. Data quality was assessed using CAMARADES, and PRISMA guidelines were followed. RESULTS:Searches identified 2175 publications from which 342 were selected for full text review, and 61 papers describing 103 experiments on 53 potential biomarkers were analyzed. The models were very heterogeneous and a meaningful meta-analysis was not possible for any biomarker. However, a pragmatic list of biomarkers of potential value in AMI that could warrant investigation in humans has been formulated. Those proposed are D-lactate, intestinal fatty acid-binding protein, ischaemia-modified albumin, and signal peptide-CUB-EGF domain-containing protein 1. CONCLUSIONS:No single biomarker has provided unambiguous evidence to promote its use in clinical practice, but several may deserve further evaluation in future human studies. SYSTEMATIC REVIEW REGISTRATION:PROSPERO CRD42023400888 (reference 400888).
BACKGROUND:Digital biomarkers-objective, quantifiable physiological and behavioral measures collected through digital tools-enable continuous health assessment and hold promise for precision medicine. With non-communicable diseases (NCDs) being the leading cause of morbidity and mortality worldwide, digital biomarkers can provide early detection, continuous monitoring, and individualized care. METHODS:This review examines advances in digital biomarkers across NCD domains, including endocrinology, cardiology, respiratory medicine, neurology, and mental health, with a focus on wearable technologies that continuously capture real-world behavior and physiology. RESULTS:Continuous glucose monitoring in diabetes exemplifies successful clinical translation, while digital biomarkers in cardiology, respiratory medicine, and neurology are at varying stages of validation and adoption. Most digital biomarkers across internal medicine, neurology, and mental health remain in early validation stages. Critical barriers to implementation include limited validation in diverse populations, lack of interoperability, insufficient integration with electronic health records, challenges in multimodal data synthesis, and underdeveloped regulatory pathways. Equity concerns persist as infrastructure, affordability, and capacity-building needs vary globally. CONCLUSIONS:Bridging the gap between consumer self-tracking and clinically validated, guideline-compatible digital biomarkers requires rigorous multicenter validation, standardized interfaces, open data models, secure and ethical data infrastructures, and equitable design. Coordinated efforts addressing these challenges are essential to enable digital biomarkers to improve prevention, diagnosis, and management of NCDs.
CONTEXT:The levels of alpha-1-antitrypsin (AAT) peptides could be used as prognostic biomarkers in critically ill patients, and peptide treatment has therapeutic efficiency in experimental sepsis. However, the regulation of AAT peptides in M. musculus is unknown, although they may be prerequisites for peptide-based therapies. This study aims to quantify murine AAT peptides and investigate their regulation during experimental sepsis. METHODS:LC-MS/MS was used to quantify six AAT peptides from two murine AAT isoforms in plasma of septic mice. RESULTS:Peptide mC36-2 is the most abundant and significantly regulated murine AAT peptide, with approximately 20-fold higher level than in critically ill patients. In contrast, the upregulation of peptide C42 in septic patients is absent in mice; thus, it may represent a specific marker of human disease. Nevertheless, the increase and kinetics of mC36 during experimental sepsis reflect our observations in patients. CONCLUSIONS:In summary, the conserved upregulation of AAT peptides reflects important pathophysiological mechanisms of severe infections; thus, peptide levels can be used as predictive biomarkers in experimental and clinical sepsis. Although species-specific differences exist and need to be considered, quantification of AAT peptides might help to characterize individual immune responses and build the foundation for novel therapeutic approaches in pre-clinical and clinical research.
BACKGROUND:The scientific community continues to debate the relationship between pregnancy depression and oxidative stress marker changes. The study tracked oxidative stress marker development in pregnant women with depression from the beginning to the end of their pregnancy. METHOD:Sixty-two pregnant women were included in the cross-sectional study. According to the Beck Depression Scale, the pregnant women were divided into depressed (n = 23, score ≥ 10) and control (n = 39, score < 10) groups. Catalase (CAT), superoxide dismutase (SOD), and malondialdehyde (MDA) levels were measured in blood samples. RESULTS:The Beck score was 17.83 ± 6.86 in pregnant women with depression and 4.51 ± 2.54 in the control group (p < 0.001). There was no difference between the groups in baseline CAT, SOD, and MDA levels (p > 0.05). MDA levels in pregnant women with depression increased significantly in the third trimester compared to the second trimester (from 4.02 ± 0.49 to 6.19 ± 0.52 nmol/mL, p < 0.01). A positive correlation was found between body mass index and MDA (r = 0.300, p < 0.05). CONCLUSION:The progression of pregnancy leads to increased lipid peroxidation in depressed women but depression itself does not impact oxidative stress markers. The extent of oxidative damage increases when mothers become obese during pregnancy.
BACKGROUND:Myocardial injury is a serious complication of acute carbon monoxide poisoning in elderly patients, we aimed to evaluate its association with short-term mortality and to develop a simplified prognostic model with internal validation. METHODS:This retrospective cohort study with 90-day follow-up included consecutive patients aged ≥ 65 years with acute carbon monoxide poisoning (2019-2024). Myocardial injury was defined as cardiac troponin I ≥ 0.05 ng/mL. Analyses included non-parametric tests, survival analysis, and ROC curves. RESULTS:Among 258 patients, 70 had myocardial injury. These patients were older and had higher rates of renal dysfunction, hospitalization, and severe neurological impairment, along with worse biochemical profiles (lower pH and base excess, reduced PaCO2, higher lactate). The 90-day mortality was significantly higher in the injury group (24.3% vs. 2.1%, P < 0.001). Myocardial injury was the strongest independent predictor of death (adjusted OR 8.16, 95% CI: 2.49-26.78), followed by neurological impairment and age. A model combining these three predictors showed good discriminative ability (AUC 0.879, 95% CI: 0.793-0.964). CONCLUSION:Myocardial injury is a key prognostic factor in elderly patients with carbon monoxide poisoning. A model incorporating myocardial injury, neurological status, and age may support risk stratification, pending external validation.
Hepatocellular carcinoma (HCC) develops within a complex microenvironment in which cancer-associated fibroblasts (CAFs) are abundant, yet their heterogeneity and functional roles remain incompletely understood. Here, we integrated multi-site single-cell RNA sequencing of Normal liver, primary Tumor, portal vein tumor thrombus (PVTT), and metastatic Lymph node samples with bulk transcriptomic analysis, cell-cell communication inference, and functional assays to define stromal programs associated with HCC progression. We identified four transcriptionally distinct CAF subsets: SGCA+ mCAFs, POSTN+ mCAFs, TDO2+ CAFs, and CXCL14+ iCAFs, each showing site-specific distribution and distinct prognostic relevance. Among them, POSTN+ mCAFs were enriched for extracellular matrix organization and integrin signaling, and their signatures were consistently associated with poor overall survival across independent cohorts. Ligand-receptor analysis identified POSTN+ mCAFs as highly connected stromal hubs communicating with malignant and myeloid cells through ECM-integrin, collagen-SDC4, and chemokine pathways. Consistently, tumor-derived POSTN-high fibroblasts promoted hepatoma cell proliferation, migration, and invasion in vitro. Together, these findings identify POSTN+ mCAFs as a transcriptionally distinct and functionally pro-tumorigenic stromal subset in HCC, supporting POSTN-centered stromal programs as candidate markers of aggressive tumor microenvironments and a basis for future mechanistic investigation.
BACKGROUND:Although systemic inflammation is a well-established prognostic factor in malignancies, composite inflammation-based models remain underexplored in Hodgkin lymphoma (HL). We investigated the prognostic significance of complete blood count (CBC)-derived inflammatory indices and a novel Complete Blood Count Score Model (CBCSM) in patients with HL. METHODS:This study included 80 patients diagnosed with HL. Neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), systemic immune-inflammation index (SII), and systemic inflammation response index (SIRI) were calculated at diagnosis. Optimal cut-off values for progression-free survival (PFS) and overall survival (OS) were determined using receiver operating characteristic analysis. Survival outcomes were assessed using Kaplan-Meier analysis. RESULTS:Higher NLR, PLR, SII, and SIRI values were significantly associated with shorter PFS and OS (p < 0.01). While clinicopathological factors age, disease stage, and B symptoms predominantly influenced overall survival, CBC-derived inflammatory indices were also associated with both PFS and OS in survival analyses. The composite CBCSM enabled additional risk stratification, with progressively worse outcomes observed across increasing risk categories. CONCLUSION:Baseline CBC-derived inflammatory indices and the composite CBCSM may provide clinically meaningful prognostic information in HL. Their simplicity, cost-effectiveness, and routine availability make them potentially useful adjuncts to traditional prognostic factors for risk stratification and individualized patient management.
BACKGROUND:Total Antioxidant Capacity (TAC) reflects the overall antioxidants in the body, which typically decrease in response to organochlorine pesticide (OCP) exposure, due to an increase in oxidative stress; however, some studies show contrasting results. PURPOSE:Owing to the inconsistent findings reported from the studies concerning the association of the TAC in OCP-exposed individuals, a meta-analysis is conducted. METHODS:A systematic literature survey was conducted through Scopus, MEDLINE, and Web of Science, up to November 2025, to find relevant articles following the PRISMA guideline.Results: The search resulted in 11 articles, with 08 clinical and 04 exposure-based case-control experiments. The pooled results showed a significant negative association of TAC with OCPs (p < 0.01; standardized mean difference of -1.68). Egger's regression revealed no significant publication bias among the studies. The sensitivity analysis encompassing all the studies indicated the absence of any significant changes in the statistics. CONCLUSION:Our findings indicated that the concentration of TAC was significantly lower in OCP exposed cases, as compared to the controls and/or unexposed counterparts. This suggests that TAC may serve as a potential biomarker of oxidative stress associated with organochlorine pesticide exposure. Also, this analysis calls for additional studies that can strengthen this association.
BACKGROUND:Acute myeloid leukemia (AML) is a clonal malignancy characterized by impaired hematopoietic differentiation. CD36 is a transmembrane glycoprotein receptor for phospholipids, lipoproteins, and long-chain fatty acids, and has been implicated in chemoresistance. In this AML cohort we evaluated CD36 expression and its association with febrile neutropenia (FEN) and overall survival (OS) as a potential biomarker of infectious risk. PURPOSE:To assess whether CD36 expression may be associated with FEN and OS in AML. METHODS:We conducted a retrospective single-center cohort study of patients with AML, evaluating flow cytometric CD36 expression and its associations with FEN and OS during follow-up. RESULTS:Among 41 AML patients (median age, 61 years; 46.3% male), CD36 expression was <20% in 20 (48.8%) and ≥20% in 21 (51.2%). FEN occurred in 34 patients (82.9%), including 20/21 (95.2%) in the CD36 ≥ 20% group (p = 0.032). Patients with CD36 expression ≥20% also exhibited shorter OS and a higher hazard of death on univariable Cox regression (HR 3.161, 95% CI 1.276-7.827; p = 0.013). CONCLUSION:Higher CD36 expression in AML may be associated with more frequent FEN and reduced OS, suggesting its potential clinical relevance as a biomarker of infection-related vulnerability and adverse prognosis.
BACKGROUND:Selenophosphate synthetase 2 (SEPHS2) activates selenophosphate, a donor for selenocysteine required for glutathione peroxidase (GPX) production. GPX protects placental tissue. Dysregulation of SEPHS2 and reduced GPX activity may contribute to oxidative stress-mediated placental dysfunction in recurrent pregnancy loss (RPL). AIM:To evaluate SEPHS2 gene expression and plasma GPX levels as prognostic biomarkers in women with idiopathic RPL and their association with oxidative imbalance. METHODS:A total of 100 women with idiopathic RPL and 100 healthy women without history of RPL were included. SEPHS2 gene expression was quantified using quantitative reverse transcription polymerase chain reaction (qRT-PCR), and plasma GPX levels were measured using enzyme-linked immunosorbent assay (ELISA). RESULTS:SEPHS2 expression was significantly elevated in the RPL group compared with controls (mean fold change: 14 ± 11.4 vs. 5 ± 8), while plasma GPx levels were markedly reduced in women with RPL (5.57 ± 0.40 vs. 9.53 ± 3.28; P < 0.001, 95% CI: 3.32-4.62). These findings indicate an association between increased SEPHS2 expression, reduced antioxidant activity, and recurrent pregnancy loss. CONCLUSION:Elevated SEPHS2 expression suggests disrupted selenoprotein-mediated antioxidant defense in RPL. SEPHS2 and GPX may serve as prognostic biomarkers.
ContextThis study evaluated changes in salivary tobacco-specific nitrosamines (TSNAs) in smokers and non-smokers before and after Spain's smoke-free legislation.MethodsWe conducted a longitudinal study of 272 adults in Barcelona, surveyed in 2004-2005 and 2013-2014, spanning the 2006 legislation and its 2011 extension. Saliva samples were analyzed for NNN, NNK, and NNAL, and geometric means with 95% confidence intervals were calculated. Data were stratified by sociodemographics, smoking status, smoking behavior, and anthropometrics. Linear mixed-effects models assessed adjusted percentage changes in TSNA concentrations.ResultsAt baseline, overall TSNA concentrations were 1.5 [1.2-1.8] pg/mL for NNN, 1.4 [1.3-1.5] pg/mL for NNK, and 0.5 [0.5-0.6] pg/mL for NNAL. Among baseline smokers who quit, TSNAs decreased by -90.5% (NNN), -48.7% (NNK), and -86.2% (NNAL). Continuing smokers showed a 149.8% increase in NNN, while non-smokers showed no significant change.ConclusionsConcentrations of salivary TSNAs declined in those who quit smoking but increased for continuing smokers. Monitoring TSNAs in saliva is useful for assessing legislative impact.
BACKGROUND:Radiation-induced intestinal injury (RIII) is a serious complication of radiotherapy and is closely associated with mitochondrial dysfunction, but the underlying mechanism remains unclear. METHODS:Mitochondria‑related differentially expressed genes (DEGs) were identified using online databases in an RIII model. Histopathological staining analysis was performed to evaluate injury severity. The expression of Mrm2 was examined in vivo after irradiation. RESULTS:The core functional modules of the identified DEGs included mitochondrial‑coding genes and apoptosis‑related genes. Histopathological analysis showed that RIII aggravated significantly in a dose‑ and time‑dependent manner. Notably, Mrm2 was continuously upregulated after irradiation in vivo. CONCLUSION:Mrm2 may serve as a key regulatory factor mediating mitochondrial dysfunction in radiation‑induced intestinal injury.
A precise diagnosis and customized treatment become more difficult by the genomic heterogeneity of breast cancer (BRCA). In order to examine gene expression data from two separate Gene Expression Omnibus (GEO) microarray datasets, we used a integrative approach in this study that combined bioinformatics and machine learning. We were able to distinguish between universal and subtype-specific transcriptome patterns by identifying both common and subtype-specific differentially expressed genes (DEGs) using dual-level differential expression analysis. Functional enrichment analysis and the creation of protein-protein interaction networks identified important hub genes, including TPM3, MYLK, and COL17A1, which showed substantial dysregulation and were linked to high mutation rates and a bad prognosis. Survival analyses, which identified COL17A1 as a predictive predictor for the general population and MYLK for the Luminal B subtype, highlighted the clinical significance of these hub genes. We used both Random Forest and K-Nearest Neighbors classifiers to ensure robust biomarker identification. In the analysis, we prioritized 35 model-agnostic biomarkers that performed well in subtype categorization, such as PNMT and KRTAP10-8. This dual-model approach improved the reliability of biomarker identification while reducing model-specific biases. These results set the stage for early identification, more accurate subtype classification, and possible therapeutic targeting in breast cancer.
