This research investigates how IDH-mutant gliomas transform from slow-growing tumors into aggressive, fatal malignancies. By comparing single-cell gene expression and chromatin accessibility, the authors found that these tumors start as indolent cells resembling oligodendrocyte progenitors (OPCs) fueled by epigenetic changes like DNA hypermethylation. As the disease progresses, the tumors shift toward a proliferative state resembling neural progenitor cells (NPCs) that share a similar epigenetic landscape but enact different genetic programs. This transition is marked by a loss of global DNA methylation, forcing the cancer to acquire new genetic mutations to maintain growth and suppress immune responses. Specifically, the study highlights how interferon signaling is initially silenced by methylation in low-grade cases but is later disrupted by genetic deletions in high-grade cases. These insights suggest that targeting epigenetic mechanisms may be most effective early in the disease before genetic alterations become the primary drivers.

References:

• Wu J, Gonzalez Castro L N, Battaglia S, et al. Evolving cell states and oncogenic drivers during the progression of IDH-mutant gliomas[J]. Nature Cancer, 2025, 6(1): 145-157.

This research article from Nature Cancer details a comprehensive study of olfactory neuroblastoma (ONB) using single-cell RNA sequencing to map its complex cellular landscape. By analyzing thousands of individual cells, researchers identified intra-tumoral heterogeneity and established a new tripartite classification system consisting of neural, basal, and mesenchymal subtypes. The study reveals that these distinct molecular profiles originate from different lineages within the olfactory epithelium, such as globose and horizontal basal cells. These findings are clinically significant as they offer biomarkers for more accurate diagnosis and suggest that specific chemotherapy regimens can be tailored to a tumor's subtype. Furthermore, the analysis of the tumor microenvironment points toward potential new immunotherapy targets involving tumor-associated macrophages. Ultimately, this work provides a foundational framework for precision medicine strategies to treat this rare and aggressive malignancy.

References:

• Yang J, Song X, Zhang H, et al. Single-cell transcriptomic landscape deciphers olfactory neuroblastoma subtypes and intra-tumoral heterogeneity[J]. Nature Cancer, 2024, 5(12): 1919-1939.

This research investigates how hepatocellular carcinoma (HCC) persists as minimal residual disease following treatment and eventually causes cancer recurrence. By utilizing spatial analysis and high-resolution imaging on human samples and mouse models, the authors identified a specific neighborhood of cells where immunosuppressive macrophages interact with stem-like tumor cells. These interactions are driven by the TGFβ pathway, which allows cancer cells to evade the immune system and leads to the exhaustion of protective CD8+ T cells. The study demonstrates that dual-targeting of the PD-L1 and TGFβ pathways can successfully eliminate these residual cells in mice. Ultimately, these findings suggest a new therapeutic strategy to prevent liver cancer from returning by disrupting the local environment that sustains dormant tumor cells.

References:

• Lemaitre L, Adeniji N, Suresh A, et al. Spatial analysis reveals targetable macrophage-mediated mechanisms of immune evasion in hepatocellular carcinoma minimal residual disease[J]. Nature cancer, 2024, 5(10): 1534-1556.

This research presents a comprehensive pan-cancer resource consisting of 1,294 plasma cell-free DNA samples to improve noninvasive tumor detection. By developing a standardized computational workflow, the authors integrated DNA methylation and fragmentomic data across eleven distinct cancer types. The study identified thousands of cancer-associated markers while effectively mitigating technical biases and biological confounders like age and sex. Their findings demonstrate that combining methylation with fragmentation features—such as 5′ end motifs and fragment ratios—significantly enhances the accuracy of cancer classification. This large-scale compendium serves as a foundational dataset for advancing liquid biopsy technologies and monitoring disease evolution. Success was further confirmed through independent validation, proving the robustness of these integrated biomarkers for clinical use.

References:

• Zeng Y, Abelman D D, Singhawansa A, et al. A pan-cancer compendium of 1,294 plasma cell-free DNA methylomes and fragmentomes enabling multicancer detection[J]. Nature Cancer, 2026: 1-15.

This research article introduces a high-throughput epitranscriptomic screening platform designed to identify functional m6A RNA modifications that drive cancer progression. By utilizing a Cas13-directed methyltransferase system, the researchers mapped the functional landscape of these modifications in prostate and lung cancer models, uncovering over 200 sites that regulate cell growth. A primary discovery identifies CHD9 as a critical tumor-suppressive gene whose protein levels are enhanced by m6A deposition at a specific site. Mechanistically, the study reveals that m6A readers YTHDF1 and YTHDF3 promote the translation of CHD9, which subsequently anchors the protein MYBBP1A in the nucleoplasm to activate p21-mediated growth arrest. Ultimately, this work provides a scalable framework for systematic RNA-level editing and highlights a novel regulatory axis for controlling tumor suppression.

References:

• Xu X, Wang Y, Zhu H, et al. METTL3-based epitranscriptomic editing screening identifies functional m6A sites in cancers[J]. Nature Cancer, 2026: 1-15.

This research identifies CRTAM as a critical immune-related adverse event (irAE) checkpoint that governs the balance between immunotherapy efficacy and side effects. By analyzing patient data and mouse models, the authors discovered that CRTAM+ T cells drive inflammatory toxicity in normal organs like the lungs and colon without significantly affecting antitumor immunity. This specific toxicity occurs because CRTAM interacts with CADM1, a protein found more abundantly in healthy epithelial tissues than in malignant tumors. Experiments showed that inhibiting or deleting CRTAM reduces severe inflammation and tissue damage while preserving the "hot" tumor environment necessary for cancer treatment. Furthermore, the study suggests that the CRTAM–type 3 immune axis can serve as a biomarker to monitor patients for potential side effects. Ultimately, targeting CRTAM offers a promising strategy to decouple treatment toxicity from efficacy, allowing for safer and more effective cancer immunotherapy.

References:

• Ma S C, Rong Z X, Xu Z P, et al. CRTAM inhibition mitigates toxicity of immune checkpoint inhibitors without antitumor efficacy trade-off[J]. Nature Cancer, 2026: 1-17.

The paper introduces GENEVA, a scalable single-cell resolution platform designed to improve the accuracy of preclinical cancer drug testing. By creating mosaic tumor models that blend diverse patient-derived and cancer cell lines into single experiments, the platform captures how different genetic backgrounds respond to treatments simultaneously. Applying this technology to KRAS-G12C inhibitors, researchers discovered that mitochondrial hyperactivation drives cell death while epithelial-to-mesenchymal transition acts as a specific resistance mechanism in living models. These high-resolution insights allowed for the identification of novel drug combinations, such as pairing KRAS inhibitors with mTOR or TGFβ/EMT blockers, to enhance therapeutic efficacy. Ultimately, the framework aims to bridge the gap between initial laboratory findings and actual patient outcomes by accounting for the inherent complexity of human tumors.

References:

• Yu J X, Suh J M, Popova K D, et al. The GENEVA platform models tumor mosaicism to reveal variations of responses to KRAS inhibitors and identify improved drug combinations[J]. Nature Cancer, 2026: 1-16.

The paper describes a study published in Nature Cancer that utilized multiplexed ion beam imaging (MIBI) and longitudinal profiling to identify biomarkers for immunotherapy response in metastatic triple-negative breast cancer. By analyzing 103 patients from the TONIC trial, researchers developed SpaceCat, an open-source pipeline that extracts over 800 spatial features from the tumor microenvironment. The study found that on-treatment metastatic biopsies were significantly more predictive of patient outcomes than primary tumors or baseline samples. High-performing multivariate models revealed that immune diversity, T cell infiltration at tumor borders, and PDL1 expression on myeloid cells are critical indicators of response. Ultimately, the research emphasizes that spatiotemporal dynamics are essential for understanding how metastatic lesions evolve and respond to immune checkpoint inhibition.

References:

• Greenwald N F, Nederlof I, Sowers C, et al. Temporal and spatial composition of the tumor microenvironment predicts response to immune checkpoint inhibition in metastatic TNBC[J]. Nature cancer, 2026: 1-16.

This research report evaluates the SC291 hypoimmune (HIP) CD19 CAR T cell product, an allogeneic therapy engineered to treat cancers and autoimmune diseases without triggering patient rejection. By utilizing CRISPR-Cas editing to remove HLA markers and overexpress CD47, these specialized cells successfully bypassed both innate and adaptive immune responses in clinical trial participants. The study observed that while patients' immune systems attacked partially edited cells, the fully edited HIP CAR T cells remained protected and functionally active. Furthermore, researchers discovered that a lack of donor-specific antibodies against unedited cells serves as a reliable indicator for deep tissue B cell depletion. These findings confirm the HIP platform's ability to evade allorejection, offering a promising path toward universal, off-the-shelf cell therapies. Ultimately, the data support the reliability of this engineering concept in overcoming the significant immunological barriers that typically limit allogeneic treatments.

References:

• Hu X, Beauchesne P, Wang C, et al. Hypoimmune CD19 CAR T cells evade allorejection in patients with cancer and autoimmune disease[J]. Cell Stem Cell, 2025, 32(9): 1356-1368. e4.

This research details the creation of a massive pancreatic cancer organoid biobank used to investigate therapeutic resistance and personalized medicine. By performing comprehensive multi-omics profiling on 260 organoid lines, scientists identified novel genetic drivers in both coding and noncoding DNA regions that influence tumor growth. A key discovery revealed that chemoresistant tumors are characterized by elevated levels of protein glycosylation and cholesterol metabolism. The study demonstrates that statins can effectively reverse this resistance by blocking these metabolic pathways and inhibiting the epithelial-to-mesenchymal transition. These laboratory findings were further validated in a phase 2 clinical trial, where adding a statin to standard chemotherapy significantly reduced tumor markers in over 70% of patients with advanced pancreatic cancer. This integrated approach offers a promising framework for identifying new biomarkers and improving treatment outcomes for one of the most lethal forms of cancer.

References:

• Li Y, Tang S, Wang H, et al. A pancreatic cancer organoid biobank links multi-omics signatures to therapeutic response and clinical evaluation of statin combination therapy[J]. Cell Stem Cell, 2025, 32(9): 1369-1389. e14.