Basic Sciences

Introduction Advanced primary liver cancers, including hepatocellular carcinoma and cholangiocarcinoma, remain difficult to treat once they become refractory to standard systemic therapies. Survival outcomes are poor, and therapeutic options are limited, particularly in heavily pretreated patients. With the growing understanding of tumor biology, comprehensive genomic profiling has emerged as a promising strategy to identify actionable molecular targets and guide personalized therapy, but its real-world clinical utility in liver cancers has remained uncertain. Problem Statement Although next-generation sequencing is increasingly used, standard panels often fail to detect a significant proportion of actionable alterations. Moreover, it is unclear whether identifying such alterations truly translates into meaningful clinical benefit, especially in patients who have already failed multiple lines of therapy. Summary This multicenter French initiative demonstrates that comprehensive genomic profiling using whole-genome, whole-exome, and RNA sequencing is feasible in advanced liver cancers and identifies actionable alterations in nearly two-thirds of patients. Importantly, this approach detects more targets than conventional sequencing panels. Matched targeted therapies based on ESCAT I–III alterations led to meaningful disease control in a subset of patients, particularly in cholangiocarcinoma and combined tumors, with significantly improved progression-free survival. In contrast, no benefit was observed when therapies were guided by lower-evidence ESCAT IV alterations. These findings highlight that the clinical value of precision oncology depends not only on identifying mutations but also on the level of evidence supporting their actionability. Early genomic profiling may allow better patient selection, preserve liver function, and expand access to effective targeted therapies in advanced hepatobiliary cancers.

Introduction Cholestatic liver disease (CLD) is a progressive condition characterised by impaired bile formation and flow, leading to bile accumulation, liver injury, and eventual cirrhosis or liver failure. Despite advances in hepatology, treatment options remain limited, with ursodeoxycholic acid and obeticholic acid offering only partial benefit. Increasing attention has been directed toward the gut–liver axis, particularly the role of gut microbiota and their metabolites in modulating liver disease. Problem Statement The key mechanisms driving CLD progression remain incompletely understood, and current therapies fail to adequately halt disease progression in many patients. Although berberine has shown hepatoprotective effects, its very low systemic bioavailability raises uncertainty about how it exerts clinical benefit, suggesting an indirect mechanism possibly mediated through gut microbiota. Summary This study demonstrates that orally administered berberine is converted by gut microbiota into dihydroberberine, which acts as the active metabolite. Dihydroberberine suppresses serotonin (5-HT) production in intestinal enterochromaffin cells by inhibiting tryptophan hydroxylase 1, thereby disrupting the 5-HT signalling axis involved in cholestatic injury. This mechanism was consistently validated across multiple animal models and further supported by clinical data showing improved biochemical markers and reduced 5-HT levels in patients. These findings highlight a novel microbiota-driven therapeutic pathway, positioning berberine as a promising candidate in CLD management.

Obesity is no longer viewed only as a metabolic disorder; it is now a major cancer-promoting state. This review explains how excess adiposity drives cancer through intertwined biologic pathways including chronic inflammation, hormonal dysregulation, immune suppression, altered energy metabolism, DNA damage, and gut microbiome disruption. For clinicians, the key message is practical: obesity is a modifiable cancer risk factor, and meaningful weight loss may reduce future cancer burden. Main clinical message Overweight and obesity are associated with higher rates of multiple cancers, especially endometrial, colorectal, liver, gallbladder, pancreas, kidney, postmenopausal breast, oesophagal adenocarcinoma, ovarian, thyroid, gastric, prostate, and multiple myeloma. The review estimates that obesity contributes to about 10% of new cancers annually in the US, and even more in selected tumour types such as endometrial and hepatobiliary malignancies. Key biologic pathways The review highlights 5 major mechanisms:

1. Adipose tissue dysfunction: enlarged adipocytes produce excess estrogens, leptin, inflammatory cytokines, and less adiponectin.
2. Chronic inflammation: IL-6, TNF-α, prostaglandin E2, and related mediators create a pro-tumor microenvironment.
3. Immune escape: obesity impairs cytotoxic T cells and NK cells while increasing immunosuppressive myeloid-derived suppressor cells.
4. Metabolic support for tumors: adipose tissue supplies free fatty acids and other fuel for cancer growth.
5. DNA damage and microbiome changes: oxidative stress and dysbiosis increase genomic instability and mucosal inflammation. Important epidemiologic insights Cancer risk is not determined by BMI alone. Patients with metabolically unhealthy obesity appear to have the highest cancer risk. The review also stresses that childhood and adolescent obesity trajectories may influence cancer risk later in life. Interestingly, obesity increases postmenopausal breast cancer risk, but may show a different association before menopause. Weight loss and cancer prevention The review suggests that modest weight loss may not be enough. A threshold of more than 10% body weight reduction may be needed to produce measurable reductions in obesity-related cancer risk. Observational data suggest benefit with: Bariatric surgery, especially for endometrial cancer risk reduction GLP-1 receptor agonists, with retrospective data suggesting lower incidence of some obesity-related cancers Metformin and related metabolic therapies, though stronger prospective evidence is still needed Practice implications Clinicians should view obesity management as part of long-term cancer prevention, not only cardiovascular and metabolic risk reduction. Counselling should move beyond BMI to include metabolic health, waist circumference, adiposity pattern, and sustained weight-loss strategies. Multimodal care combining lifestyle measures, pharmacotherapy, and in selected patients, bariatric surgery may have future oncologic relevance. Limitations of the review Much of the evidence linking weight loss interventions to lower cancer incidence remains observational, not randomised. Several mechanistic pathways are strongly biologically plausible but not yet fully translated into cancer prevention trials. Bottom line Obesity promotes cancer through multiple biologic pathways, and meaningful sustained weight loss may reduce this risk. This review strengthens the concept that treating obesity is also part of cancer prevention.

Introduction Gastric cancer remains one of the most lethal malignancies worldwide. The tumour microenvironment (TME)—particularly cancer-associated fibroblasts (CAFs)—plays a critical role in tumour progression, immune evasion, and resistance to immunotherapy. Although immune checkpoint inhibitors (ICIs) have improved outcomes in some patients with gastric cancer, many fail to respond due to a highly immunosuppressive TME. Understanding the specific fibroblast subpopulations responsible for immune suppression may help identify new therapeutic targets. Summary In this study, researchers used single-cell RNA sequencing and spatial transcriptomics from gastric cancer tissues to identify a distinct fibroblast subset characterized by phosphodiesterase type 5A (PDE5A) expression. Key findings include: PDE5A⁺ CAFs were associated with poorer overall survival and a strongly immunosuppressive tumour microenvironment. These fibroblasts promoted extracellular matrix remodeling and epithelial–mesenchymal transition (EMT) in gastric cancer cells. PDE5A⁺ CAFs activated the PI3K/AKT/mTOR pathway, leading to secretion of CXCL12, which interacts with CXCR4 to recruit dysfunctional CD8⁺ TEX⁺ LAG3 T cells, thereby suppressing effective anti-tumor immunity. Tumors enriched with PDE5A⁺ CAFs showed T-cell exclusion and reduced cytotoxic CD8⁺ T-cell infiltration, contributing to immunotherapy resistance. Importantly, combined therapy using a PDE5A inhibitor (vardenafil) with LAG3 immune checkpoint blockade significantly improved antitumor responses and reduced tumor growth in mouse models. Key Message PDE5A⁺ cancer-associated fibroblasts represent a critical driver of immune suppression in gastric cancer, and targeting this pathway may enhance the effectiveness of immunotherapy.

Introduction Most obesity therapies reduce body fat, but they also cause a meaningful loss of lean mass, including skeletal muscle. This matters because preserving muscle is important for physical function, metabolic health, and long-term weight loss. Bimagrumab is an investigational monoclonal antibody that blocks activin type II receptors, promoting fat loss and muscle preservation/growth. Semaglutide, a GLP-1 receptor agonist, mainly reduces weight by lowering appetite and food intake. The BELIEVE phase 2 trial tested whether combining these two drugs could produce greater weight loss with better body composition than either alone. Summary In this randomised phase 2 trial, 507 adults with obesity received placebo, bimagrumab, semaglutide, or combinations for 48 weeks, followed by extension to 72 weeks. At week 48, body weight fell by 9.3 kg with high-dose bimagrumab alone, 14.2 kg with semaglutide 2.4 mg, and 17.8 kg with the high-dose combination, versus 3.3 kg with placebo. By week 72, the high-dose combination achieved about 22.1% weight loss, greater than semaglutide alone. The major strength of the combination was body composition. It produced striking reductions in total fat mass and visceral fat while preserving lean mass far better than semaglutide alone. At week 72, the high-dose combination reduced fat mass by 45.7% and visceral adipose tissue by 58.2%, while limiting lean-mass loss. Glycemic measures, waist circumference, hsCRP, and several metabolic markers also improved. Adverse effects reflected known profiles of both drugs. Bimagrumab was associated with muscle spasms, diarrhea, and acne, while semaglutide caused nausea, diarrhea, constipation, and fatigue. Key Message This study suggests that bimagrumab plus semaglutide may offer a next-generation obesity treatment approach: substantial weight loss with enhanced fat loss and relative preservation of muscle mass.

Chromosome 8q amplification, a common genomic alteration in hepatocellular carcinoma (HCC), includes the gene TATA-box binding protein–associated factor 2 (TAF2), a key component of the TFIID basal transcription complex. This study demonstrates that TAF2 is overexpressed in human HCC tissues and cell lines, and higher expression correlates with poorer overall survival. Using hepatocyte-specific TAF2 knockout mice, investigators showed that loss of TAF2 leads to hepatocyte death, compensatory proliferation, inflammation, and fibrosis, creating a microenvironment favorable for tumor development. Consequently, DEN/high-fat high-sugar diet–induced HCC was markedly increased in TAF2-deficient mice. Conversely, TAF2 overexpression alone did not initiate liver tumors, but significantly enhanced MYC-driven hepatocarcinogenesis, suggesting that TAF2 acts as a tumor promoter rather than a primary oncogenic driver. Mechanistically, TAF2 binds promoter regions of tumor-promoting genes and non-coding RNAs, enhancing transcription programs that support cancer hallmarks such as proliferation and survival. Key Message: TAF2 plays a central role in maintaining hepatocyte viability and can accelerate hepatocarcinogenesis in the presence of oncogenic signals, highlighting it as a potential biomarker and therapeutic target in HCC.

Introduction Understanding which genes restrain gastric tumor growth—and how host factors like Helicobacter pylori shape tumor biology—remains central to precision prevention and therapy. CRISPR-Cas9 loss-of-function screening offers a scalable way to discover tumor suppressors, but most screens are performed in vitro and miss key in vivo pressures such as immunity and microbial influences. This study builds an organoid-based in vivo CRISPR platform to identify gastric tumor suppressors in both ectopic (subcutaneous) and orthotopic (stomach) settings. Summary (≈200 words) Using murine gastric organoids, the authors performed in vivo CRISPR knockout screening with (1) a custom library targeting 49 putative gastric tumor suppressors and (2) a genome-scale “cancer” library targeting ~5000 genes. Screens were conducted across immunocompetent and immunodeficient mice, with and without H pylori infection, and in both subcutaneous and surgically implanted orthotopic tumor models. Recurrently enriched guides identified Pten, Fbxw7, and multiple TGF-β pathway components (Smad4, Tgfbr1, Tgfbr2, Acvr2a) as consistent tumor suppressor hits across models; genome-scale screening confirmed these and revealed additional candidates. The top hits were individually validated in vivo. Mechanistically, Pten loss drove large, highly vascular tumors with neutrophil recruitment and T-cell exclusion, highlighting an immune-evasive, pro-angiogenic state. In contrast, loss of Smad4, Tgfbr1, or Acvr2a produced lesions resembling early gastric precancer states, including Alcian blue–positive intestinal metaplasia and compensatory hyperplasia. Notably, H pylori did not change the tumor mutational landscape; instead, it primarily reshaped the tumor microenvironment, promoting influx of tumour-supporting SiglecF⁺ neutrophils. Overall, the work introduces a versatile in vivo organoid-CRISPR platform that separates tumor genetics from host factors while capturing clinically relevant gastric cancer biology.

Introduction Most clinicians recognise classic inherited polyposis syndromes such as familial adenomatous polyposis (FAP) from pathogenic variants in APC, and MUTYH-associated polyposis (MAP) from biallelic MUTYH variants. Yet, a large proportion of patients with 20–100 adenomas (and even some with hundreds) remain “genetically unexplained” after routine multigene panel testing. This editorial discusses why that gap persists and why newer molecular approaches are beginning to close it. Summary Llor reviews major advances that explain previously unresolved polyposis cases. A key theme is that many APC pathogenic alterations are missed by standard testing, including deep intronic variants causing aberrant splicing or pseudoexon formation, synonymous changes that disrupt splicing, large structural rearrangements, and mobile element insertions—often requiring whole-exome/genome sequencing, long-read sequencing, and RNA-based analyses for detection. Another important mechanism is somatic mosaicism, where post-zygotic APC mutations are present only in a subset of tissues; phenotype depends on when the mutation occurs in development and can appear attenuated in the proband but become more severe in offspring if transmitted through germline involvement. Mosaic APC variants may account for a meaningful fraction of “unexplained” FAP-like presentations. The editorial highlights new work by Sommer et al. in this issue, who studied individuals with unexplained adenomatous or serrated polyposis and found APC mosaicism in ~19% of evaluable adenomatous polyposis cases, with additional suspected mosaicism. The broader implication is practical: when multigene panels are negative, clinicians and genetic services should increasingly consider expanded diagnostics that include intronic/regulatory APC alterations and mosaicism testing. Finally, Llor notes that serrated lesions show distinct molecular patterns (often BRAF), but their biology may not mirror adenomatous APC-driven progression; many serrated lesions may represent benign clonal outgrowths requiring additional hits to progress—an area where major mechanistic questions remain.

Introduction UGT1A1 poor metabolisers (PMs) are at higher risk of severe irinotecan-related toxicity. A 30% upfront dose reduction is commonly recommended for safety, but whether this compromises survival has remained uncertain. Summary In this Dutch multicentre retrospective cohort (2017–2024), including 779 patients with colorectal or pancreatic cancer, 9.8% were UGT1A1 poor metabolisers who received a 30% reduced irinotecan dose. Progression-free and overall survival were comparable between dose-reduced PMs and fully dosed intermediate/normal metabolisers. Severe toxicity rates were also similar. These findings suggest that genotype-guided dose reduction improves safety without compromising survival, supporting routine UGT1A1-guided irinotecan dosing in clinical practice.