BRAFV600-mutant metastatic colorectal cancer comprises a biologically distinct and clinically aggressive subset of metastatic colorectal cancer. Early attempts to apply single-agent BRAFV600 inhibition failed because of rapid acquired resistance and reactivation of mitogen-activated protein kinase signaling. Over the last decade, rational combinations (BRAF inhibitors and epidermal growth factor receptor inhibitors with or without mitogen-activated protein kinase kinase inhibitors) have become the standard of care in the refractory setting and are now being evaluated upfront, whereas a wave of next-generation approaches (extracellular signal-regulated kinase and SHP2 inhibitors, receptor tyrosine kinase-targeted agents, and immunotherapy combinations) aims to prevent or overcome resistance. The objective of this comprehensive review was to summarize historic therapeutic approaches, current standards, and the mechanistic rationale and clinical development of next-generation strategies to combat resistance in BRAFV600-mutant metastatic colorectal cancer.

Tumors exhibiting mismatch repair deficiency without detectable germline mutations via standard multigene panel testing are often classified as Lynch-like syndrome. In the present report, we describe the case of a 47-year-old man presenting with synchronous axillary sebaceous carcinoma and colonic medullary carcinoma. Although initial germline multigene panel testing failed to identify pathogenic variants, a high degree of pathological suspicion for Muir-Torre syndrome remained. Immunohistochemistry revealed a concordant loss of MSH2 and MSH6 expression in both the extraocular sebaceous carcinoma and the colonic medullary carcinoma. This identical protein-loss pattern detected across anatomically distinct tumors served as decisive pathological evidence of an underlying germline defect rather than independent biallelic somatic mutations. Consequently, whole-genome sequencing was performed to resolve the discrepancy between the pathological findings and multigene panel testing results, successfully identifying a germline intronic MSH2 variant (NM_000251.3:c.2459-12A>G). Subsequent RNA analysis confirmed aberrant splicing with an 11-bp insertion. In conclusion, the present case illustrates that concordant mismatch repair protein loss across multiple tumors provides compelling morphological evidence to guide pathologists in reconsidering negative panel results and pursuing comprehensive genomic investigation to identify pathogenic intronic variants.

DNA methylation-based classification is a key adjunct to histopathology in CNS tumor diagnostics. Although both fresh-frozen and FFPE tissues are used, the impact of sampling strategy on classifier performance remains insufficiently characterized. We analyzed 179 CNS specimens (162 tumors and 17 controls), including 42 FFPE and 137 fresh-frozen samples with ≥ 50% tumor cellularity. Fresh-frozen cases were stratified into histology-validated and blind sampling groups. Genome-wide DNA methylation profiling was performed using Illumina EPIC/EPIC v2.0 arrays and classified using the DKFZ Brain Tumor Classifier, with independent validation using the NIH Methylscape platform. Results were categorized as confirmed, refined, changed, or unclassified and correlated with histopathological diagnoses. FFPE samples showed high diagnostic yield with minimal unclassified results (2%). Among fresh-frozen specimens, histology-validated samples demonstrated robust classification, whereas blind sampling showed higher discordance and unclassified rates (22%). Discordant results were confined to the blind-sampling group and were most consistent with sampling bias and variable tumor representation rather than technical limitations. Methylation profiling confirmed or refined diagnoses in most cases and led to clinically relevant reclassification in a subset. Classifier performance appeared to be influenced more by sampling strategy rather than preservation method, with histology-guided sampling associated with the most reliable diagnostic outcomes.

Esophageal squamous cell carcinoma (ESCC) features epithelial heterogeneity and an immunosuppressive microenvironment, yet clinically relevant malignant epithelial states remain poorly defined. We integrated four single-cell RNA-sequencing datasets and 10 bulk transcriptomic cohorts totaling 1318 samples, and applied cross-cohort differential expression, survival analysis, and a machine-learning framework of 113 algorithm combinations to screen for malignant epithelial cell-associated biomarkers. GNGT1 was prioritized for its consistent upregulation, prognostic association, and limited prior characterization in ESCC. Across independent cohorts, GNGT1 exhibited favorable diagnostic performance, while high expression correlated with poorer overall survival and more advanced local tumor status. Immune deconvolution consistently linked GNGT1-high tumors to reduced CD8+ T-cell infiltration, lower immune scores, and decreased cytotoxic T-cell markers. Single-cell analyses further associated GNGT1-high epithelial cells with altered epithelial-immune communication involving MIF, prostaglandin, and CXCL signaling, alongside enrichment of epithelial-mesenchymal transition, mTORC1, and proliferative programs; qPCR confirmed elevated GNGT1 expression in ESCC cell lines. This study defines GNGT1 as a marker of an immunosuppressive malignant epithelial state, thereby bridging epithelial heterogeneity with immune remodeling in ESCC. These findings support GNGT1 as a candidate diagnostic, prognostic, and biologically informative biomarker warranting mechanistic and clinical validation.

Lysine β-hydroxybutyrylation (Kbhb) is a metabolite-derived post-translational modification of histone and non-histone proteins that couples β-hydroxybutyrate (BHB) availability to gene expression. Yet the transcription factors that govern the Kbhb substrate program in cancer remain unidentified. Existing studies have cataloged Kbhb-modified substrates or examined individual proteins, without identifying the transcriptional regulators of the program in a defined tumor context. Here, we performed network-based master regulator analysis (MRA), implemented in the viper package, on a molecular signature restricted to experimentally validated Kbhb substrates, across two independent PAM50 Basal-like breast cancer (BLBC) cohorts profiled on orthogonal platforms: TCGA-BRCA (RNA-seq; n = 195 tumor, 113 normal) and METABRIC (microarray; n = 209 tumor, 148 normal). Dataset-specific regulatory networks were inferred with ARACNe-AP and integrated by cross-platform Stouffer meta-analysis. Of 1493 Kbhb substrates, 1322 and 1213 were expressed in the respective cohorts. The analysis identified seven concordant transcriptional master regulators (six activated, one repressed; cross-cohort NES correlation r = 0.64), with CENPA (meta-NES +4.55) and FOXM1 (meta-NES +4.27) as the dominant drivers. These findings nominate a BHB-Kbhb-FOXM1/CENPA axis linking ketone-body metabolism to mitotic transcription, with potentially protumoral implications for ketogenic regimens in BLBC.

Exosomes are small extracellular vesicles that mediate intercellular communication and, in cancer, carry cargo that both reflects the donor tumor cell and influences recipient cells within local and distant microenvironments. Exosomal proteins and microRNAs have been reported individually across many cancer types, but rarely compared on a common basis; in this review, previously reported molecules from eight cancer categories-blood, breast, colon, kidney, liver, lung, prostate, and stomach-were compiled from curated repositories and re-analyzed within a single functional framework. In total, 3643 exosomal proteins (523 hematologic, 3120 solid-tumor) and 627,225 miRNA-target pairs, derived from 350 unique microRNAs, were organized using Gene Ontology, KEGG, and PANTHER annotation. Across cancers, proteins converged on a reproducible core-signaling, transport, cytoskeletal organization, and extracellular interaction-dominated by binding, catalytic, and transporter functions localized to membrane, vesicle, and extracellular compartments. Comparisons between hematologic and solid malignancies revealed both shared cancer-associated functions and context-dependent patterns linked to tissue origin and disease ecology. Together, these findings indicate that integrated protein-and-microRNA profiling offers a useful framework for understanding tumor communication, refining cancer classification, and advancing biomarker discovery, while underscoring that harmonized workflows, independent validation, and mechanistic follow-up remain necessary before descriptive enrichment outputs can support clinically robust applications.

As an important member of the ADAMTS (a disintegrin and metalloproteinase with thrombospondin motifs) family, ADAMTS16 participates in various physiological and pathological processes through the regulation of extracellular matrix (ECM) components. This review systematically delineates the expression profiles of ADAMTS16 in different tissues and cell types, as well as its diverse roles in reproductive development, cardiovascular diseases, ocular diseases, skeletal disorders, malignancies, and other pathological conditions, revealing significant differences in its expression patterns and mechanisms of action across various tissues. Cumulatively, current evidence suggests that ADAMTS16 may engage in direct or indirect interactions with ECM constituents to regulate cell proliferation, differentiation, and migration, and may also influence a range of cellular activities via synergistic cooperation with other cytokines.

The primary aim of this review is to summarize current evidence from clinical and pre-clinical studies on endogenous molecular biomarkers associated with radiosensitivity and radioresistance in cervical cancer, including patients treated with photon-based radiotherapy, cervical cancer cell lines, and xenograft models, and to evaluate the association of these biomarkers with radiotherapy response, residual disease, recurrence and survival, and experimental measures of radiosensitivity. A systematic literature review was conducted for studies published over the last 10 years that evaluated associations between genomic, epigenetic, or protein biomarkers and radiotherapy response or survival outcomes in cervical cancer. Eligible studies included in the current analysis summarize clinical, translational, and pre-clinical studies in correlation with photon-based radiotherapy. Research focusing exclusively on non-coding RNAs, exogenous radiosensitizers, or non-photon modalities was excluded. In total, 112 studies were identified, and 46 of them met the inclusion criteria. The identified biomarkers clustered into several key biological processes: DNA damage response and cell cycle regulation, cancer stemness, hypoxia and microenvironment, epigenetic and transcriptional regulation, and signaling pathways, including exosome-mediated communication. Most markers were linked to radioresistance and adverse outcomes, whereas a smaller subset was associated with increased radiosensitivity. A limited group of biomarkers was linked to clinical outcomes such as local control, residual disease, or survival, and emerging multi-marker protein signatures suggested that combinatorial approaches may outperform single-marker strategies. Radiosensitivity in cervical cancer is regulated by a network of biological pathways. Validated, integrated biomarker panels that capture DNA repair proficiency, stemness, hypoxia adaptation, and key signaling pathways are needed to improve risk stratification and enable biomarker-guided radiosensitization.

Identifying immunotherapy biomarkers for colorectal cancer (CRC) beyond MSI-high status is challenging. We evaluated CREBBP/EP300 and POLE/POLD1 mutations as candidate markers using public cohorts (MSK-CHORD, n = 5493; TCGA, n = 528; MetTropism, n = 24,496; ICI-treated, n = 1610). Regression with SHAP attribution, survival analyses, and pan-cancer transcriptomic profiling (n = 7628) were performed. CREBBP/EP300 mutations were independently associated with TMB-high MSS tumors (p < 0.00001). The co-mutation group (CREBBP/EP300 + POLE/POLD1; CoMut) was the strongest predictor of elevated TMB level (β = 3.63; p < 10-16; exp(β) = 37.7) with a median TMB of 155.9 vs. 7.4 and 5.4 Mut/Mb in single-mutation groups. CoMut also associated with improved survival (HR = 0.57, p = 0.035). In the ICI-treated cohort, both single-mutation groups (CREBBP/EP300-MUT-only; POLE/POLD1-MUT-only) and showed longer overall survival than wild type (34 and 28 vs. 17 months; p < 0.05), but only CREBBP/EP300 mutations independently reduced mortality risk (HR = 0.71, p = 0.0025). All groups shared immune-inflammatory transcriptomic enrichment. These findings support CREBBP/EP300 mutations as candidate immunotherapy biomarkers, and co-mutations with POLE/POLD1 define a TMB-high MSS subgroup that could refine TMB testing, pending prospective validation.

This study assessed the prevalence and clinical relevance of p16INK4a, p53, HER2, and Nectin-4 protein expression in primary vaginal cancer (PVC), with HER2 and Nectin-4 representing potentially druggable targets. A total of 34 patients with PVC from the University Medical Center Mainz biobank were included. Immunohistochemistry was used to evaluate the expression of p16, p53, HER2, and Nectin-4: p16, p53, and HER2 staining results were dichotomized (positive vs. negative; mutant vs. wildtype staining pattern), whereas Nectin-4 expression was assessed using an immunoreactive score (IRS) based on staining intensity and the proportion of positive tumor cells. Univariate analyses were performed to identify intergroup differences, and survival analyses were conducted using the log-rank test. Negative p16 expression was significantly associated with adenocarcinomatous histology (p = 0.0211), but neither p16 nor p53 status showed prognostic relevance. HER2 positivity was detected in 14.7% of tumors and was significantly associated with adenocarcinoma (p = 0.0001), p16 negativity (p = 0.0035), and younger age (p = 0.0483). Based on IRS, strong total Nectin-4 expression was observed in 20.59% and moderate membranous Nectin-4 expression in 17.65% of cases. No significant associations were found between HER2 or Nectin-4 expression and clinical outcomes; however, a limited statistical power of the survival analysis due to the low number of events (n = 10 deaths) should be considered. This analysis demonstrates a notable prevalence of HER2 and Nectin-4 expression in PVC.

Cancer pain affects more than half of all oncology patients and represents one of the most challenging therapeutic problems in modern medicine. Despite the central role of opioids in cancer pain management-codified by the WHO analgesic ladder-the clinical response to these agents is highly variable between individuals. Pharmacogenomics promises to explain this heterogeneity by identifying functionally relevant genetic polymorphisms in genes encoding opioid receptors, metabolizing enzymes, and membrane transport proteins. However, after two decades of intensive research, clinical translation of genetic data into actionable prescribing strategies remains elusive. A narrative-critical review of the literature was conducted by searching PubMed, Scopus, and Web of Science databases from inception to March 2026, using the search terms "opioid," "pharmacogenomics," "pharmacogenetics," "cancer pain," "polymorphism," and individual gene names (OPRM1, CYP2D6, COMT, ABCB1, SLC22A1, UGT2B7, and others). Reference lists of included articles were hand-searched for additional relevant studies. Inclusion was restricted to English-language articles reporting original data or systematic reviews on genetic polymorphisms and opioid response, with priority given to studies conducted in cancer pain populations. Clinical practice guidelines from CPIC, NCCN, ESMO, EAPC, and ASCO were consulted for current recommendations. This narrative-critical review examines the evidence on genetic polymorphisms relevant to opioid response in cancer pain, covering three functional categories: (1) pharmacodynamic genes, including opioid receptors (OPRM1, OPRD1, OPRK1) and neuromodulatory targets (COMT, KCNJ6, MC1R); (2) pharmacokinetic enzymes (CYP2D6, CYP3A4/5, UGT2B7); and (3) membrane transporters, including ABC efflux pumps (ABCB1, ABCG2, ABCC2) and SLC uptake carriers (SLC22A1, SLCO1B1, SLC6A4, SLC6A2). A dedicated section addresses the direct and indirect interference of the tumour itself-through the tumour microenvironment, neuroimmune signalling, epigenetic reprogramming, and cancer-induced organ dysfunction-on opioid pharmacogenomics. The overall evidence is poor and fragmented, a finding intrinsic to the extraordinary diversity, large number, and context-dependent biological activity of the polymorphisms identified. The tumour itself acts as a pervasive confounder that systematically distorts genotype-phenotype relationships. The path forward requires a paradigm shift from candidate-gene association studies to system pharmacogenomics and multi-omic integration.

Homer proteins (Homer1-3) are scaffold proteins that mediate protein-protein interactions in signal transduction; however, the role of Homer3 in breast cancer (BC), particularly triple-negative breast cancer (TNBC), remains poorly defined. Here, we investigated the clinical relevance and functional significance of Homer3 in BC and TNBC. Publicly available datasets from TCGA and GEO were analyzed to evaluate associations between Homer3 expression and patient outcomes using Kaplan-Meier survival analysis. Pathway enrichment analysis and gene set variation analysis (GSVA) were performed to identify signaling programs associated with Homer3 co-expressed genes. Functional roles were examined using stable TNBC cell lines with Homer3 knockdown or overexpression, followed by assays for cell proliferation, clonogenic growth, migration, invasion, and wound healing. We found that Homer3 expression was elevated in breast tumors compared with normal tissues and was associated with poor prognosis in both BC and TNBC. Homer3 expression was higher in TNBC than in non-TNBC subtypes and negatively correlated with estrogen receptor (ER) and progesterone receptor (PR) expression. Functionally, Homer3 depletion suppressed TNBC cell proliferation, clonogenic capacity, migration, invasion, and wound healing, whereas Homer3 overexpression produced reciprocal effects. Pathway analyses revealed that Homer3 co-expressed genes were enriched in cell cycle-related pathways and Hallmark MYC signaling, which were associated with adverse clinical outcomes. Consistently, Homer3 knockdown selectively reduced key proliferative cell-cycle regulators. Collectively, these findings demonstrate that Homer3 is associated with aggressive phenotypes and MYC- and cell cycle-linked proliferative programs in breast cancer, particularly in TNBC. Survival analyses are presented as exploratory findings, supporting the biological relevance of Homer3 in TNBC. These findings suggest that Homer3 may represent a potential therapeutic vulnerability.

Adrenocortical carcinoma (ACC) is a rare malignancy of the adrenal gland for which no curative treatment options exist in advanced stages. Analyzing the ACC cohort of The Cancer Genome Atlas, we identified that expression of the mRNA-stabilizing protein Embryonic-Lethal-Abnormal-Vision-Like RNA-Binding Protein 1 (ELAVL1) was negatively associated with patient survival. To investigate the functional role in ACC, we generated a CRISPR/Cas9-mediated ELAVL1 knockout (KO) in the ACC cell lines NCI-H295R and HAC15 and performed Next-Generation RNA Sequencing (NGS) to characterize transcriptomic changes. In NCI-H295R, NGS analysis revealed 3468 upregulated genes and 3458 downregulated genes in ELAVL1-deficient cells compared with controls. Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis identified significant enrichment of aldosterone biosynthesis-related genes and enriched downregulation of genes associated in with "pathways in cancer." Functionally, ELAVL1-KO cells exhibited impaired colony-forming ability, indicating reduced proliferative or clonogenic potential. Steroid profiling of cell culture supernatants further demonstrated increased aldosterone synthesis and decreased cortisol and androgen production in the KO cells, consistent with the observed transcriptional changes. Given that hypercortisolism is an established negative prognostic factor in ACC, these data suggest that ELAVL1 may represent a potential therapeutic target worth further investigation for modulating steroidogenesis in ACC patients.

RNA-binding proteins (RBPs) remain underexplored as small-molecule targets, although their dysregulation contributes to numerous human diseases, including cancer. RBPs are key regulators of post-transcriptional gene expression, controlling multiple stages of RNA metabolism. Among them, insulin-like growth factor 2 mRNA-binding protein 3 (IGF2BP3) is an oncofetal RBP that is highly expressed during embryonic development, largely absent in adult tissues, and re-expressed in multiple malignancies. A growing body of evidence supports IGF2BP3 as a diagnostic and prognostic biomarker and a potent oncogenic driver across tumor types, underscoring its potential as a therapeutic target. However, the development of effective IGF2BP3-targeting compounds remains in its early stages. In this review, we first describe the structural organization of IGF2BP3, the molecular basis of RNA recognition, and the mechanisms underlying its dysregulation across human cancers. We then discuss emerging therapeutic approaches, including direct inhibition of IGF2BP3-RNA interactions and indirect strategies that rewire IGF2BP3 expression or activity through epigenetic, epitranscriptomic, and signaling pathways. By critically highlighting the opportunities and limitations of these approaches and their impact on cancer progression, we provide an integrated perspective combining structural biology, medicinal chemistry, and cancer biology to support the development of next-generation IGF2BP3-targeted therapies.

Here, we report the identification of a previously unrecognized population of tumor-induced podoplanin-positive (PDPN+) cells in clear-cell renal cell carcinoma (ccRCC) that exhibit features of under-differentiated lymphatic endothelial cells (LECs). These PDPN+ cells lack a complete repertoire of canonical LEC markers, including VE-cadherin, LYVE1, and VEGFR3, and fail to form functional lymphatic vessels, indicating a dysplastic phenotype. We termed these cells dysLECs and found that these cells are induced by BMP10 produced specifically by tumor cells. In turn, dysLECs secrete CXCL13, which promotes tumor cell proliferation and metastasis. Ligand-receptor analyses revealed a highly tumor-specific reciprocal signaling circuit: kidney tubule cells deficient in the von Hippel-Lindau (VHL) tumor suppressor gene uniquely express BMP10, a TGF-β family cytokine, whereas its receptor ALK1 is restricted to dysLECs; conversely, dysLECs produce CXCL13, while VHL mutant kidney tubule cells uniquely express its receptor, CXCR5. Pharmacological inhibition of ALK1 reduced CXCL13 production and suppressed the hyperplastic phenotype of VHL mutant tumor cells in vivo, whereas BMP10 neutralization inhibited tumor growth and metastasis in an orthotopic ccRCC xenograft model. Collectively, these findings identify a dysplastic population of PDPN+ lymphatic-like endothelial cells and define a tumor-specific BMP10-CXCL13 signaling axis that drives ccRCC progression, uncovering a previously unrecognized therapeutic vulnerability in this disease.

Pancreatic ductal adenocarcinoma (PDAC) has two principal molecular subtypes-classical and basal-like-with divergent prognosis and chemotherapy response, yet transcriptomic classifiers rarely generalize across cohorts or quantify per-patient uncertainty. We trained a classical-versus-basal-like classifier on CPTAC-PDAC (n = 140) and externally validated it on histology-filtered TCGA-PAAD (n = 150). Twelve algorithms were benchmarked under stratified nested cross-validation with four-method consensus feature selection; domain adaptation (naive transfer, CORAL, and ComBat), four conformal procedures (split, weighted, CV+, and Conformal Risk Control), and a four-method consensus explainable-AI framework (SHAP, LIME, permutation importance, and decision-curve ablation) with pathway enrichment were then evaluated. Top models reached a cross-validated AUROC ≈ 0.96 and external AUROC 0.913-0.938 (top-3 ensemble 0.961); batch correction did not improve transfer, indicating minimal residual batch effect. Consensus explainability recovered keratinization biology and nominated five candidate genes (GSDMC, A2ML1, PIP5K1B, IL20RB, and AKR7L) beyond the Moffitt signature. All four conformal procedures plateaued near 0.85 coverage at α = 0.05 under zero-shot transfer, whereas local recalibration on a small target sample restored nominal coverage. We present a transparent, externally validated, uncertainty-aware and TRIPOD+AI-compliant PDAC subtype classifier, best deployed as a calibrated decision-support tool with site-specific recalibration.

Lung cancer remains the leading cause of cancer-related mortality worldwide, and durable benefit from immune checkpoint inhibitors is limited by immune exclusion, microenvironmental immunosuppression, and toxicity. This narrative translational review evaluates whether endobronchial intratumoral delivery of immuno- and gene-based therapies can transform bronchoscopic access into a therapeutic platform linking local tumour intervention with systemic antitumour immunity. We synthesise evidence on endobronchial ultrasound, electromagnetic navigation bronchoscopy, and robotic-assisted bronchoscopy, together with local checkpoint blockade, cytokine and mRNA-lipid nanoparticle constructs, oncolytic virotherapy, dendritic-cell approaches, viral and non-viral gene transfer, and enzyme- or metabolite-based strategies. Current clinical evidence remains preliminary, consisting mainly of Phase I trials, small prospective cohorts, and case-based signals, with feasibility and safety observations but no completed randomised trial demonstrating efficacy against standard-of-care systemic therapy. The most plausible candidates are patients with advanced or recurrent NSCLC, bronchoscopically accessible lesions, inadequate response to systemic immunotherapy, and immune-excluded or locally immunosuppressed tumours. Biomarkers such as PD-L1, tumour mutational burden, CD8+ infiltration, tertiary lymphoid structures, radiomics, and circulating tumour DNA require validation. This review integrates delivery technology, immune mechanisms, statistical evidence appraisal, biomarker limitations, AI-guided planning, and global feasibility.

Glioblastoma (GBM) is a highly aggressive primary malignant brain tumor, featuring diffuse infiltrative growth and poor clinical outcomes, which underscores the need to delineate the molecular mechanisms driving its malignant progression. COP9 signalosome subunit 7B (COPS7B), a core component of the conserved COP9 signalosome complex, is significantly upregulated in GBM tissues; however, its biological function and regulatory mechanism in GBM remain largely elusive. Here, we found that elevated COPS7B expression was positively correlated with glioma pathological grade and adverse prognosis in histologically and molecularly confirmed GBM patients. Functional assays demonstrated that COPS7B markedly promoted the proliferation, migration, and invasion of GBM cells in vitro, while COPS7B knockdown exerted the opposite suppressive effects. Mechanistically, we identified clusterin (CLU) as a key downstream functional effector of COPS7B in GBM. COPS7B upregulated CLU protein abundance by enhancing the translation efficiency of CLU mRNA, without altering its transcriptional level or protein stability. Functional rescue experiments further confirmed that CLU is indispensable for COPS7B-mediated malignancy-driving phenotypes in GBM, and transcriptomic analysis revealed that the progression-promoting effect of CLU was tightly associated with the activation of tumor-related signaling cascades, including the ERK and MAPK pathways, as well as the regulation of cell growth, invasion, and migration. Collectively, this study not only reveals a critical role of COPS7B in driving GBM malignant progression but also delineates a novel COPS7B-CLU regulatory axis that drives GBM aggressive phenotypes via activation of mitogenic signaling, suggesting candidate targets for further translational investigation.

Artificial intelligence (AI) has emerged as a promising tool for inferring molecular biomarkers directly from digitized histopathologic slides. However, the current evidence in gastrointestinal and hepatobiliary cancers remains fragmented across tumor types and biomarker categories. Relevant studies were systematically identified and screened according to the PRISMA 2020 statement. PubMed, Embase, Web of Science Core Collection, Scopus, and IEEE Xplore were searched from 1 January 2015 to the date of search. Eligible studies involved gastrointestinal or hepatobiliary malignancies, used histopathology images, applied AI-based methods, and reported molecular biomarker prediction or inference. A total of 110 studies were included. Most studies focused on colorectal, gastric, liver, and pancreatic cancers, with microsatellite instability, mutation status, molecular subtypes, and tumor mutational burden being the most commonly investigated targets. Model architectures evolved from conventional convolutional neural networks to multiple-instance learning and transformer-based methods. While many studies reported promising predictive performance, direct comparison across studies remained challenging because of substantial heterogeneity in datasets, model architectures, and validation strategies. AI-based molecular biomarker identification from pathologic slides shows substantial promise in gastrointestinal and hepatobiliary cancers, but current evidence is constrained by heterogeneity and limited validation. Standardized, multicenter studies are needed before routine clinical implementation.

Non-small-cell lung cancer (NSCLC) is one of the most frequent cancer types and is responsible for the majority of cancer-related deaths worldwide. For this reason, initial diagnosis, prognosis, and targeted therapy of NSCLC represent very attractive areas of study. Aptamers are single-stranded nucleic acids (RNA or DNA) generated through Systematic Evolution of Ligands by Exponential Enrichment (SELEX); they are able to bind to a molecular target with high affinity and specificity. Thanks to their intrinsic nucleic acid properties, they can be easily modified and optimized to enhance target binding and their half-life; moreover, they exhibit no immunogenicity and toxicity. Due to this, many aptamers have just been selected against NSCLC biomarkers and provide specific imaging agents to improve the diagnosis of this type of cancer. However, despite the promising results in preclinical studies, the application of aptamers in NSCLC diagnosis is still in its early stages, mainly due to the limited literature in the research world, the dominance of antibodies in the pharmaceutical market and the challenge of target identification. Consequently, this appears to be a temporary issue, and aptamers could see increasing application in the future; thus, we performed a review aimed at summarizing current knowledge on the new promising DNA and RNA aptamer-based molecules for NSCLC diagnosis. All studies from 2000 were included and investigated. Our findings showed that several DNA and RNA aptamers are promising diagnostic tools for NSCLC management.