ClairS is a novel deep-learning method specifically designed for calling somatic small variants in tumor-normal pairs using long-read sequencing data. Unlike most existing tools optimized for short reads, ClairS was trained on synthetic somatic variants and real cancer cell lines. The method demonstrated high accuracy, achieving F1 scores of 96.19% for SNVs and 79.67% for indels after training augmentation. The authors emphasize that improved read phasing enabled by long-read sequencing is key to accurate SNV detection, especially at low variant allele fractions. ClairS is presented as a robust and reliable open-source caller.
Prognostication and clinical management of central conventional chondrosarcomas are challenging due to their heterogeneous behavior and the high interobserver variability of histological grading. This study investigated the prognostic value of DNA methylation profiling to improve risk stratification. Using methylation data from 69 primary tumors, a LASSO Cox regression model identified eight informative methylation sites. These sites were used to construct a risk score, named CHROME, which predicts disease recurrence, onset of metastasis, and disease-specific mortality. Validated in an independent cohort of 68 patients, CHROME demonstrated strong discriminatory performance, accurately stratifying patients into High or Low risk groups.
This study introduces a stacked ensemble machine learning classifier designed to predict the tissue of origin for cancers using circulating tumor DNA (ctDNA). The model integrates 11 multidimensional ctDNA features, encompassing genomic, fragmentomic, methylation/repeat, and microbial signals, derived from low-pass whole-genome sequencing. It demonstrated robust performance, achieving top-1 and top-2 accuracies of 80% and 90% in an independent validation cohort of 1,221 samples. The classifier proved effective even with low tumor fractions and successfully identified the primary site in 73.3% of cancers of unknown primary (CUP) cases. This non-invasive approach holds significant potential for guiding clinical decision-making.
This study analyzed 159 patients with secondary glioblastoma (sGBM) and confirmed that the PTPRZ1-MET fusion is a recurrent oncogenic driver associated with significantly shorter overall and progression-free survival. Researchers identified 359 upregulated genes in fusion-positive tumors, enriched in cell-cycle regulation pathways. An artificial intelligence-based diagnostic model, utilizing a three-gene panel (MET, PCDHGA3, FAM3C), was developed and validated to identify this fusion. This robust molecular signature could facilitate the identification and stratification of sGBM patients positive for the PTPRZ1-MET fusion.
This article outlines the protocol for an umbrella review aiming to synthesize evidence on prognostic biomarkers in bladder cancer. The objective is to compare all identified prognostic biomarkers in the literature to assess their clinical value. The authors plan to search systematic reviews and meta-analyses across multiple databases, extract relevant data, and assess the risk of bias. Statistical analysis will include pooled HR estimation, evaluation of heterogeneity and small-study effects, and an assessment of evidence credibility.
This multicenter retrospective Japanese study compared afatinib and osimertinib as first-line treatments for advanced or recurrent non-small cell lung cancer (NSCLC) harboring uncommon EGFR mutations. Weighted analyses revealed no significant differences in time to treatment failure or overall survival between the two EGFR-TKIs. While afatinib was associated with higher response rates and more frequent adverse events requiring dose reduction, subgroup analyses suggested differential treatment effects based on mutation subtype. Furthermore, subsequent immune checkpoint inhibitor-based regimens showed limited additional benefit.
This study aimed to develop a mutational signature-based biomarker (MSBM) to define homologous recombination deficiency (HRD) in advanced ovarian cancer, overcoming limitations of genomic scar-based assays. The MSBM integrates tumor-BRCA mutation status, Mutational Signature 3 similarity score, and the absence of CCNE1 amplification, derived from whole-exome sequencing. A phase II trial (MSBM-OL) validated this biomarker and evaluated the efficacy of olaparib maintenance monotherapy in HRD-positive patients. The study demonstrated that olaparib significantly extended progression-free survival (22.3 months) compared to a predefined 9-month threshold, confirming its clinical benefit as first-line maintenance therapy.
This article highlights the significance of rare and ultrarare gene fusions, beyond established ones like ALK and ROS1, in lung cancer. It emphasizes that emerging fusions, involving receptor tyrosine kinases or their ligands, such as EGFR-SHC1, unveil novel mechanisms of oncogenic activation. These discoveries are propelling precision medicine towards an era of "ultraprecision" oncology, enabling increasingly individualized treatment strategies for cancer patients.
This Australian multicenter study, derived from the AURORA registry, analyzed real-world treatment sequencing and survival in 115 patients with ROS1-rearranged NSCLC between 2012 and 2025. It revealed that a majority of patients (74%) received a ROS1 inhibitor as first-line therapy, with increasing use of later-generation inhibitors. Median overall survival was 56 months, extending to 80 months for those treated with a later-generation ROS1 inhibitor in the first line. Brain metastases and high PD-L1 expression were associated with shorter overall survival.
This study evaluates the efficacy of qPCR for HER2 status assessment in invasive ductal carcinoma of the breast, comparing it with IHC and FISH. Results demonstrate high concordance between qPCR and FISH, particularly in resolving equivocal IHC 2+ cases. Furthermore, exploratory genomic profiling via WES on high-risk subtypes (HER2 3+ and TNBC) revealed somatic mutations in key genes such as TP53, BRCA1, and MYCN, as well as AR expression in TNBC. These findings highlight qPCR's potential as an accurate diagnostic tool and provide a foundation for future precision oncology strategies.
This multi-omics bioinformatics study investigated the significance of TOP2A in renal cell carcinoma subtypes KIRC, KIRP, and KICH. Analyses revealed significant TOP2A overexpression associated with advanced pathological stages, high tumor grades, and poor patient outcomes. Genomic alterations, primarily amplifications and mutations, were identified at the TOP2A locus. The study suggests that TOP2A regulates the cell cycle, influences the immune microenvironment, and modulates key signaling pathways, positioning it as a prognostic biomarker and a potential therapeutic target.
This retrospective study developed machine learning models to improve long-term prognostic assessment in endometrial cancer. By integrating routine laboratory tests with FIGO staging, researchers demonstrated a significant enhancement in predicting 5- and 10-year overall and disease-free survival. The K-nearest neighbour model achieved an AUC of 0.876 for 10-year overall survival, outperforming FIGO staging and age alone. This low-cost and accessible approach could optimize prognostic assessment, especially in settings where advanced molecular profiling is limited.
This study investigates a simplified approach to identify lung origin in cancers of unknown primary (CUP). Researchers found that the presence of SMARCA4 mutations, combined with a history of smoking, is strongly associated with lung-origin CUP (CUP-Lung). This combination offers a 76% probability of lung origin, even without detectable pulmonary involvement. This method could serve as a surrogate for whole-genome sequencing (WGS) to guide organ-directed treatments.
This retrospective study compared the clinicopathological features of stage I-II oral squamous cell carcinoma (OSCC) in adolescents and young adults (AYAs) versus older patients. OSCCs in AYAs are more frequently located on the tongue and exhibit distinct histological patterns, including an abnormal TP53 immunophenotype. Although AYAs demonstrate significantly better overall and disease-free survival, a depth of invasion (DOI) greater than 5 mm is a major prognostic indicator for distant metastasis and poorer survival within this group. Other factors such as clinical stage, tumor thickness, and tumor budding may also predict the risk of postoperative lymph node metastasis. Tested molecular markers did not provide robust prognostic stratification.
This study, conducted within the INFORM program, optimized liquid biopsy methodologies for high-risk pediatric solid tumors. Researchers demonstrated that low-coverage whole-genome sequencing (lcWGS) reliably detects circulating tumor DNA (ctDNA). An in silico ctDNA estimation score, combining fragment length and genome segment alterations, improved sensitivity and specificity to 95%, enabling plasma-based tumor detection in 93% of patients. Whole-exome sequencing (WES) and targeted panel sequencing effectively identified clinically relevant, potentially druggable molecular targets, and liquid biopsy showed potential for tracking tumor evolution and refining patient stratification. These advancements lay the groundwork for integrating liquid biopsy into personalized medicine programs and pediatric clinical trials.
This study employed a bioinformatics approach to identify differentially expressed genes (DEGs) in various types of colorectal adenomas (tubular, tubulovillous, villous) compared to normal tissues. Gene expression data analysis revealed 1,024 common DEGs, with COL1A2 and CXCL8 identified as hub genes. Specific genes for each adenoma type were also discovered. RT-qPCR validation and ROC analysis confirmed the discriminative potential of CXCL8 and COL1A2 in distinguishing adenomatous/cancerous tissues from normal mucosa. These findings offer valuable insights into adenoma biology and suggest potential targets for colorectal cancer prevention and treatment.
ProphDR is a novel interpretable deep learning model designed to predict cancer drug responses by integrating multi-omics data and drug structural information. The model employs hierarchical attention mechanisms, including a Criss-Cross Gene-level Multiomics Integration (CGMI) module and a cross-attention (CA) module to model drug-gene interactions. Evaluated on GDSC and CCLE datasets, ProphDR achieves state-of-the-art performance in predicting ln(IC50) values and classifying drug sensitivity. It also demonstrates strong generalizability, even for unseen drugs or cell lines, and generates biologically interpretable attention maps.
This study investigated the mechanisms and therapeutic targets of distant metastasis in lung adenocarcinoma (LUAD), with a focus on organotropism. By integrating digital spatial profiling (DSP), multiplex immunofluorescence (mIF), and clinical data, researchers developed highly accurate random forest models predicting organ-specific metastases to the brain, liver, adrenal gland, and bone. Key compartment-specific gene expression signatures were identified in tumor, immune, and stromal cells, including FKBP1A for brain and MOCOS for liver metastasis. The study also revealed distinct biological pathways and prognostic factors related to post-metastasis survival, providing promising new biomarkers and therapeutic targets.
This study investigates how altered enhancers in prostate cancer cooperate within the 3D chromatin architecture to drive oncogenic programs. Researchers identified prostate cancer-specific enhancers that form highly nested interactions with promoters, coalescing into multi-connected hubs. CRISPR/Cas9 perturbations distinguished distinct enhancer classes: central enhancers, whose deletion collapses hub-wide activity and impairs cell proliferation, and redundant enhancers, whose deletion has minimal impact due to compensatory architectural rewiring. The transcription factor FOXA1 was found to directly bind and regulate these distinct enhancer classes. These findings suggest that enhancers function in a coordinated manner and open new avenues for precision therapies.
This phase 3 study evaluated the efficacy and safety of daraxonrasib, a multiselective RAS(ON) inhibitor, in patients with previously treated metastatic pancreatic ductal adenocarcinoma (mPDAC). Patients were randomized to receive either daraxonrasib or investigator's choice chemotherapy. Results demonstrated significantly longer overall survival and progression-free survival with daraxonrasib compared to chemotherapy, both in the RAS G12 mutated population and the overall population. Daraxonrasib also showed a favorable safety profile, with a significantly lower rate of treatment discontinuation due to adverse events compared to chemotherapy.
This study established 18 patient-derived glioblastoma organoid (GBO) lines that faithfully preserve the histopathological and genomic features of parental tumors. Glioblastoma organoid-based drug sensitivity testing (GBO-DST) demonstrated a strong correlation with progression-free survival and proved superior to MGMT methylation status in predicting temozolomide response. Transcriptomic analysis elucidated mechanisms of temozolomide resistance, including mismatch repair deficiency and elevated MGMT expression. Furthermore, FDA-approved drug screening using GBO-DST identified regorafenib and lazertinib as effective therapeutic alternatives, with lazertinib showing superior efficacy in a GBO transplantation mouse model. These findings highlight the significant potential of GBO-DST as a robust platform for precision oncology in glioblastoma.
The prospective phase 2 DUTRENEO study demonstrated that a retrospectively validated 18-gene tumor inflammation signature (TIS) failed to predict response to neoadjuvant immune checkpoint inhibitors (ICIs) in muscle-invasive bladder cancer. This outcome indicates that bulk gene-expression stratification is insufficient to enrich for responders. Single-cell spatial transcriptomic analysis revealed that response is instead governed by spatial architectures, such as CD8+ T cell proximity to cancer cells and localized checkpoint co-expression, which are invisible to bulk assays. A quantitative framework was developed, suggesting that at least 77 genes and tissue regions of at least 3-mm diameter are required to preserve predictive spatial signals.
This study conducted a comprehensive multi-omics analysis of pulmonary sarcomatoid carcinoma (PSC), a rare and aggressive lung cancer. Integrating whole-exome sequencing, transcriptomic, proteomic, and phosphoproteomic data from 86 patients, alongside single-cell RNA sequencing, researchers characterized this poorly understood disease. They identified an elevated ferroptosis suppression signature, making tumors vulnerable to targeting FTL and SLC3A2. The study also revealed distinct patient subtypes based on proteomic profiling and highlighted the role of epithelial-mesenchymal transition in tumor progression and immunotherapy resistance.