Editors’ Selections from Relevant Scientific Publications

Familial adenomatous polyposis polyp (by DataBase Center for Life Science via Wikimedia Commons)A recent study suggests that many human premalignant colorectal lesions originate from multiple genetically distinct cell populations rather than a single mutant clone. By analyzing whole genome (WGS) and whole exome sequencing data of normal colorectal mucosa, benign and dysplastic premalignant polyps, and malignant adenocarcinomas (123 samples) from six patients with familial adenomatous polyposis (FAP), Van Egeren and colleagues reconstructed early evolutionary events in colorectal tumorigenesis. They found that 40% of benign polyps and 28% of dysplastic lesions showed clear evidence of polyclonality, with divergent lineages arising early and evolving in parallel. Single-crypt WGS of polyps in two additional patients with FAP further confirmed polyclonal initiation. Moreover, distinct APC second hit mutations within a single polyp indicated divergent tumor initiation from multiple mutant clones. These results challenge the traditional monoclonal model of cancer initiation and highlight the role of tissue context and microenvironment in early lesion development.Van Egeren D, … Curtis C. Nature. 2025 Nov 25.Commensal bacteria (red) in a mouse small intestine (by The University of Chicago via Flickr)A recent study indicates that microbiome is involved in the pathogenesis of colitis-associated cancer (CAC). Using Winnie mice, a model of chronic ulcerative colitis carrying a missense mutation in the Muc2 gene, Verna and colleagues found that CAC developed under specific pathogen-free (SPF) conditions, whereas germ-free mice were protected from colitis and colon tumors, highlighting the microbiome’s essential role. Multi-omics analyses revealed distinct microbial and metabolic signatures, including alterations in sphingolipid and nucleotide metabolism, that promote the transition from colitis to CAC. Fecal microbiota transplantation from SPF Winne mice induced CAC in germ-free recipients, confirming the microbiome’s transmissible effects. These findings underscore microbiome modulation as a promising strategy for preventing inflammation-driven colorectal cancer.Verna G, … Cominelli F. J Clin Invest. 2025 Nov 6:e196712.Evolution of LUAD (by Verna et al. via Cancer Cell)A new study reveals that epithelial-proinflammatory niches drive tumorigenesis of lung adenocarcinoma (LUAD). By analyzing spatial transcriptomic profiles of human precursor lesions and LUADs, Peng and colleagues identified KRT8-high alveolar intermediate cells (KACs) as early precursors in the evolution of LUAD. These cells reside in proinflammatory niches enriched in IL-1β–IL-1R1 signaling. Such epithelial–inflammatory niches were prevalent in precursor lesions but diminished in LUADs. Studies in tobacco carcinogen-induced mouse models confirmed the conservation of KACs and inflammatory niches, and treatment with recombinant IL-1β or coculture with lung macrophages accelerated growth of KAC-rich organoids. In addition, IL-1β blockade during the precursor stage, particularly when combined with PD-1 inhibition, significantly reduced KAC abundance, suppressed pro-tumor inflammation, and inhibited LUAD development. These findings delineate a precise epithelial-immune mechanism driving early lung carcinogenesis and suggest the IL-1β–IL-1R1 axis and inflammatory niches as promising targets for cancer interception at its earliest stage.Peng F, … Kadara H. Cancer Cell. 2025 Nov 6:S1535-6108(25)00445-3.AI-assisted analysis of CT scans for Pancreatic cancer early detection (created using Gemini)A new study demonstrates that artificial intelligence (AI) can improve detection of pancreatic cancer on routine CT scans, offering a powerful approach for early cancer detection. In the multinational PANORAMA study, Alves and colleagues established a benchmark to evaluate the performance of AI and compared it with that of 68 radiologists using real-world CT imaging from 3,440 participants. The AI system achieved an area under the receiver operating characteristic curve (AUROC) of 0.92 for identifying pancreatic ductal adenocarcinoma, compared with an AUROC of 0.88 for the pooled radiologists. Notably, the AI system detected more cancers at the same specificity and produced fewer false positives at the same sensitivity than human readers. These findings underscore the potential of AI-assisted imaging as a scalable strategy for detecting pancreatic cancer at earlier, more treatable stages.Alves N, … PANORAMA consortium.Lancet Oncol. 2025 Nov 20:S1470-2045(25)00567-4.