Tumour Biol. 2025 Jan-Dec;47:14230380251410470. doi: 10.1177/14230380251410470. Epub 2025 Dec 24.

ABSTRACT

BackgroundIn recent years, the significance of sirtuins in cancer biology has become increasingly evident, but their molecular mechanisms and prognostic impacts remain elusive.ObjectiveThe present study aimed to investigate the differential expression of the sirtuin gene family across cancers and to evaluate their prognostic value.MethodsWe used various bioinformatics databases and methodologies, including Oncomine, GEPIA, OncoDB, cBioPortal, R2 Kaplan-Meier Scanner, STRING, etc., to determine the expression pattern of the sirtuin family genes, along with their mutations and prognostic values in human cancers.ResultsIn the current study, SIRT1, SIRT2, SIRT4, and SIRT5 were downregulated in lymphoma, whereas SIRT6 and SIRT7 were overexpressed. In breast cancer, SIRT3, SIRT5, and SIRT7 were overexpressed, and in terms of kidney cancer, higher expression of SIRT2, SIRT3, and SIRT5 was observed. In contrast, for leukemia, bladder, and brain cancers, most sirtuin family members showed reduced expression. We found that most mutations occurred in uterine cancer, chRCC (chromophobe renal cell carcinoma), DLBCL (diffuse large B-cell lymphoma), melanoma, pRCC (papillary renal cell carcinoma), and esophageal cancer. Moreover, we identified the relevant functional proteins through protein-protein interaction analysis to evaluate copy number alterations (CNAs) in sirtuins. The most frequent alterations were amplifications and deep deletions. Survival analysis demonstrated that SIRT1 and SIRT2 overexpression correlated with improved overall survival in low-grade glioma but predicted poorer outcomes in ovarian cancer. Downregulation of SIRT1, SIRT3, and SIRT5 was associated with better prognosis in DLBCL, while SIRT3 and SIRT4 upregulation predicted favorable survival in testicular germ cell tumors. SIRT6 overexpression was linked to favorable prognosis in esophageal carcinoma and sarcoma, while unfavorable outcomes were observed in hepatocellular carcinoma and cholangiocarcinoma. SIRT7 upregulation was significantly associated with reduced survival in esophageal, liver, and uterine cancers, but surprisingly correlated with improved outcomes in urothelial carcinoma and cervical squamous cell carcinoma.ConclusionsTogether, this multi-omics analysis reveals the correlation and prognostic values of sirtuins across multiple types of human cancers and suggests that sirtuins may serve as promising biomarkers for different cancers.

PMID:41439701 | DOI:10.1177/14230380251410470

Crit Rev Clin Lab Sci. 2025 Dec 27:1-12. doi: 10.1080/10408363.2025.2606357. Online ahead of print.

ABSTRACT

Circulating plasma DNA has found important applications in diverse medical fields, including prenatal testing, transplantation, and especially cancer. Many companies have developed products for detecting minimal residual disease, selecting or monitoring therapy, assessing prognosis, and confirming diagnosis. One major application is in screening asymptomatic individuals for the presence of cancer. Screening may facilitate better clinical outcomes through earlier interventions. Collectively, these technologies are widely known as "liquid biopsies". After the extraction of free DNA from the circulation, it is analyzed by various molecular techniques to explore differences between DNA originating from normal cells and cancer cells. Circulating plasma DNA originating from tumors (ctDNA) is expected to harbor the same molecular changes as tumor tissue itself. Thus, ctDNA is considered a surrogate of cancer tissue, but without the need to perform invasive biopsies to obtain it. Many new diagnostic companies have taken advantage of this new biomarker and developed technologies for screening for one or multiple cancers. We previously estimated the amount of ctDNA in circulation, which is admixed with DNA originating from normal cells. We concluded that since only a small fraction of the whole plasma (3 liters) is used for testing (3 to 4 mL), it is possible that the retrieved ctDNA may not be enough for cancer diagnosis in all patients. This problem is more acute with small tumors. Here, we mention some companies in the "liquid biopsy" arena and analyze their clinical data to establish if their tests are close to entering the clinic. We conclude from this analysis that current data do not support the use of these technologies for population screening due to many false negative and false positive results.

PMID:41454842 | DOI:10.1080/10408363.2025.2606357

BMC Med Res Methodol. 2025 Dec 17. doi: 10.1186/s12874-025-02713-z. Online ahead of print.

ABSTRACT

BACKGROUND: Early cancer diagnosis is essential for improving prognosis and guiding treatment. However, the high dimensionality and complexity of omics data present major challenges. Computational approaches that extract stable biomarkers and enable reliable classification across cancer types and stages are needed.

METHODS: A novel feature selection method, sDCFE (synergistic Discriminative Cluster-based Feature Extraction), was developed by extending Fisher-like variance analysis with a median absolute deviation (MAD) regularization term and a cluster separation component to enhance robustness and interpretability. Features selected by sDCFE were compared with those obtained from XGBoost, and the intersected set of 82 genes was evaluated through functional enrichment (KEGG, Reactome, GO BP), survival analysis (Kaplan-Meier, Cox regression), and biomarker novelty assessment against six external resources. Hybrid classification models integrating XGBoost, sDCFE, and deep learning were applied to pancancer classification, and the framework was further extended to lung squamous cell carcinoma (LUSC) staging using RNA-seq and methylation data.

RESULTS: The overlap between sDCFE and XGBoost yielded 82 candidate biomarkers enriched in cancer-related pathways, including cell cycle regulation, immune signalling, and DNA repair. Novelty assessment stratified these genes into established, emerging, and novel categories. Six genes-HFE2, LOC339674, SERINC2, SFTA3, SOX2OT, and ACPP-emerged as the most promising candidates, supported by enrichment and survival associations across multiple cancers. The hybrid model achieved near-perfect pancancer classification on TCGA (accuracy = 99.3%, MCC = 0.992, AUC = 1.0) and demonstrated strong generalizability on PCAWG (accuracy = 94%, MCC = 0.929, AUC = 0.997). In the LUSC staging task, multiomics integration improved classification performance: the CNN-based model reached 84% accuracy, while logistic regression applied to sDCFE-ranked features achieved 88.5% accuracy with superior calibration, highlighting the robustness of the selected features.

CONCLUSION: sDCFE provides a principled extension of Fisher-like methods, enabling stable and interpretable biomarker selection. When combined with XGBoost and deep learning, the framework achieves highly accurate and biologically grounded cancer classification across both cancer types and stages. The identification of novel and prognostic biomarkers, including HFE2, LOC339674, SERINC2, SFTA3, SOX2OT, and ACPP, underscores its translational potential. These results position the framework as a promising precision oncology tool to support early diagnosis, risk stratification, and treatment decision-making.

PMID:41408184 | DOI:10.1186/s12874-025-02713-z