Metabolites. 2025 Aug 28;15(9):575. doi: 10.3390/metabo15090575.

ABSTRACT

INTRODUCTION: Hepatocellular carcinoma (HCC), the most prevalent form of liver cancer, ranks as the third leading cause of mortality globally. Patients diagnosed with HCC exhibit a dismal prognosis mostly due to the emergence of symptoms in the advanced stages of the disease. Moreover, conventional biomarkers demonstrate insufficient efficacy in the early detection of HCC, hence highlighting the need for the identification of novel and more effective biomarkers.

METHODS: In this paper, we investigate methods for integration of multi-omics data we generated by both untargeted and targeted mass spectrometric analysis of serum samples from HCC cases and patients with liver cirrhosis. Specifically, the performances of several feature selection methods are evaluated on their abilities to identify a panel of multi-omics features that distinguish HCC cases from cirrhotic controls.

RESULTS: The integrative analysis identified key molecules associated with liver including such as leucine and isoleucine as well as SERPINA1, which is involved in LXR/RXR Activation and Acute Response signaling. A new method that uses recursive feature selection in conjunction with a transformer-based deep learning model as an estimator led to more promising results compared to other deep learning methods that perform disease classification and feature selection sequentially.

CONCLUSIONS: The findings in this study reinforce the importance of adapting or extending deep learning models to support robust feature selection, especially for integration of multi-omics data with limited sample size to avoid the risk of overfitting and the need for evaluation of the multi-omics features discovered in this study via blood samples from a larger and independent cohort to identify robust biomarkers for HCC.

PMID:41002959 | PMC:PMC12471784 | DOI:10.3390/metabo15090575

Sci Adv. 2025 Sep 12;11(37):eady0080. doi: 10.1126/sciadv.ady0080. Epub 2025 Sep 10.

ABSTRACT

Cell type-specific regulatory programs that drive type 1 diabetes (T1D) in the pancreas are poorly understood. Here, we performed single-nucleus multiomics and spatial transcriptomics in up to 32 nondiabetic (ND), autoantibody-positive (AAB⁺), and T1D pancreas donors. Genomic profiles from 853,005 cells mapped to 12 pancreatic cell types, including multiple exocrine subtypes. β, Acinar, and other cell types, and related cellular niches, had altered abundance and gene activity in T1D progression, including distinct pathways altered in AAB⁺ compared to T1D. We identified epigenomic drivers of gene activity in T1D and AAB⁺ which, combined with genetic association, revealed causal pathways of T1D risk including antigen presentation in β cells. Last, single-cell and spatial profiles together revealed widespread changes in cell-cell signaling in T1D including signals affecting β cell regulation. Overall, these results revealed drivers of T1D in the pancreas, which form the basis for therapeutic targets for disease prevention.

PMID:40929272 | PMC:PMC12422192 | DOI:10.1126/sciadv.ady0080

Eur J Radiol. 2025 Jul 22;191:112327. doi: 10.1016/j.ejrad.2025.112327. Online ahead of print.

ABSTRACT

The NAVIGATOR project established an Italian regional imaging biobank and interactive research platform designed to support precision oncology through the integration of multimodal imaging, clinical, and omics data. The platform goes beyond a static repository, offering a secure Virtual Research Environment (VRE) where users can upload data, test AI algorithms, and execute complete analytical pipelines. The platform incorporates artificial intelligence (AI)-driven radiomics and deep learning methodologies to enable biomarker extraction, disease stratification, and predictive modeling. This manuscript presents the development and implementation of the NAVIGATOR infrastructure, including its data governance framework, ethical and legal considerations, and application to three oncological use cases: prostate, rectal, and gastric cancers. To date, the biobank has collected imaging and clinical data from over 700 patients across these cohorts. AI models were deployed within a dedicated VRE to facilitate image analysis, feature extraction, and classification tasks. The project addresses critical challenges related to data harmonization, regulatory compliance, privacy safeguards and fairness in AI systems. NAVIGATOR demonstrates the feasibility of integrating AI methodologies within imaging biobanks and provides a scalable framework to advance oncological research and support clinical decision-making.

PMID:40743874 | DOI:10.1016/j.ejrad.2025.112327

J Liq Biopsy. 2025 Jul 16;9:100312. doi: 10.1016/j.jlb.2025.100312. eCollection 2025 Sep.

ABSTRACT

Breast cancer (BC) is the most frequent cancer and the leading cause of cancer-related death among women worldwide. It represents a heterogeneous group of diseases with distinct morphological, immunophenotypic, and molecular profiles, which significantly impact clinical behavior and therapeutic response. Moreover, under treatment pressure, tumor cells may undergo molecular changes and phenotypic plasticity, leading to resistance and therapeutic failure. Although tissue biopsy remains the gold standard for diagnosis and molecular characterization, it has several limitations, including invasiveness, sampling bias, and the inability to dynamically capture tumor evolution over time. Hence, a non-invasive and repeatable approach capable of real-time monitoring is increasingly needed. Liquid biopsy (LB), through the analysis of circulating tumor cells (CTCs) and circulating tumor DNA (ctDNA), has emerged as a powerful tool to complement tissue biopsy. It allows for longitudinal assessment of tumor burden, detection of minimal residual disease, and identification of molecular alterations relevant to targeted therapies. Despite promising results, the integration of LB into clinical practice is still limited by methodological heterogeneity, standardization gaps, and regulatory issues. Nonetheless, LB represents a key advancement toward precision oncology and may become essential in the personalized management of BC patients. In this review, we explore the current applications, benefits, and technical limitations of LB in different BC settings. We provide a comprehensive overview of the biological and clinical significance of CTCs and ctDNA, emphasizing their diagnostic, prognostic, and predictive roles. Finally, we present an updated summary of ongoing clinical trials that incorporate LB for clinical decision-making.

PMID:40740670 | PMC:PMC12308030 | DOI:10.1016/j.jlb.2025.100312

Diabetologia. 2025 Jul 30. doi: 10.1007/s00125-025-06462-y. Online ahead of print.

ABSTRACT

This article summarises the current understanding of the heterogeneity of type 1 diabetes from a June 2024 international Expert Forum organised by the editors of Diabetes, Diabetes Care, and Diabetologia. The Forum reviewed key factors contributing to the development and progression of type 1 diabetes and outlined specific, high-priority research questions. Knowledge gaps were identified and, notably, opportunities to harness disease heterogeneity to develop personalised therapies were outlined. Herein, we summarise our discussions and review the heterogeneity of genetic risk and immunologic and metabolic phenotypes that influence and characterise type 1 diabetes progression (presented as a palette of risk factors). We discuss how these age-related factors determine disease aggressiveness (along gradients) and describe how variable immunogenetic pathways aggregate (into networks) to affect beta cell and other pancreatic pathologies to cause clinical disease at different ages and with variable severity (described as disease-related thresholds). Heterogeneity of pathogenesis and clinical severity opens avenues to prevention and intervention, including the potential of disease-modifying immunotherapy and islet cell replacement. We conclude with a call for (1) continued research to identify more factors contributing to the disease, both overall and in specific subgroups; (2) investigations focusing on both individuals who surpass metabolic and immune thresholds and develop diabetes and those who remain disease free with the same level of immunogenetic risk; and (3) efforts to identify where the current type 1 diabetes staging system may fall short and determine how it can be improved to capture and leverage heterogeneity in prevention and intervention strategies.

PMID:40736750 | DOI:10.1007/s00125-025-06462-y