Nature Medicine, Published online: 28 August 2025; doi:10.1038/s41591-025-03900-7
Trained and validated on multimodal data from 14.5 million images from multicountry datasets, a foundation model is shown to increase diagnostic and referral accuracy of clinicians when used as an assistant in a trial involving 16 ophthalmologists and 668 patients.
Int J Surg. 2025 Jul 22. doi: 10.1097/JS9.0000000000002999. Online ahead of print.
ABSTRACT
BACKGROUND: Breast cancer remains the most prevalent malignancy among women, and patients presenting with both breast and lung cancer pose significant challenges in clinical diagnosis and treatment. Currently, comprehensive multi-omics analyses for such multiple malignancies are lacking.
METHODS: An integrated multi-omics analysis was performed, incorporating quantitative proteomics and radiomics data from patients with single primary breast cancer as well as those with multiple primary tumors (breast and lung cancer).
RESULTS: Quantitative proteomics analysis revealed four distinct molecular signatures (Types I-IV). Patients with single breast cancer exhibited driving pathways primarily linked to cell proliferation (e.g., HER2), whereas those with multiple breast cancers showed enrichment in ER-related and proliferative pathways. In contrast, patients with multiple lung cancers displayed pathways associated with immune response and immune escape. Additionally, immune subtyping identified three distinct immune landscapes (Types I-III). Radiomic analysis demonstrated strong correlations between these molecular/immune subtypes and imaging findings. Patients with high imaging information scores exhibited pronounced tumor heterogeneity and reduced immune infiltration.
CONCLUSIONS: This study provides new insights into the molecular pathogenesis of multiple primary malignancies, particularly breast and lung cancer.
PMID:40694032 | DOI:10.1097/JS9.0000000000002999
Login