Tumor biology and anticancer therapy: from mechanism papers to pivotal readouts and launches.
In the first clinical study of HRS-4642 combined with adebrelimab in previously treated patients with metastatic KRAS G12D-mutant pancreatic cancer, the chemotherapy-free regimen showed manageable safety and preliminary antitumor activity.
Why it matters: First clinical study of a chemotherapy-free combination in this patient population.
A single-cell atlas of bone marrow T cells from patients with bone marrow-resident malignancies reveals tumor-reactive populations in a state of latent competence. A 15-gene signature identifies this subset, which is expanded by immunotherapies and predicts clinical responses. The work pinpoints a definable T cell state that clinicians could use to gauge who will respond to immunotherapy.
Why it matters: A 15-gene signature of latent tumor-reactive T cells may predict immunotherapy response in marrow-resident cancers.
An immature CD16-CD161- double-negative cord blood NK subset drives CAR NK product dysfunction: after CAR engineering these cells remained hypofunctional yet showed exaggerated trogocytosis, forming a trogocytic-antigen sink that diverted CAR engagement from tumors and caused exhaustion, fratricide and death of CD16+CD161+ double-positive effectors. Depleting the DN subset before manufacturing eliminated the antigen sink and restored antitumor activity in hematologic and solid tumor models. The work converts donor heterogeneity into an actionable manufacturing step for more potent off-the-shelf CAR NK products.
Why it matters: It identifies a donor subset-driven mechanism of CAR NK dysfunction and a simple pre-manufacturing depletion that restores potency.
The clinically used hypomethylating agent 5-azacytidine kills leukemia cells by depleting 5-methylcytosine in chromatin-associated RNA, not only through DNA demethylation. This caRNA m5C loss disrupts MBD6-mediated H2AK119ub deubiquitination, impairs SRSF2 recruitment and p300-driven H3K27ac deposition, and represses genes for cell-cycle regulation and DNA repair. Loss of the caRNA methyltransferase NSUN2 produced a prolonged cell cycle, defective DNA repair and a shift toward erythropoiesis, mirroring 5-azaC's effects.
Why it matters: It identifies RNA m5C depletion in chromatin-associated RNA as a mechanism of 5-azacytidine action in leukemia.
Personalized cancer vaccines, delivered as synthetic long peptide or DNA formulations after resection and adjuvant chemotherapy, were safe and immunogenic in pancreatic ductal adenocarcinoma: two phase 1 trials showed no grade ≥3 adverse events and induced polyclonal neoantigen-specific T cell responses confirmed by IFN-γ ELISpot, intracellular cytokine staining, and T cell receptor clonotype reconstruction. Compared with a propensity-matched cohort, vaccine recipients showed a trend toward longer median overall survival (4.4 versus 3.5 years, log-rank P = 0.23). The results establish feasibility of neoantigen vaccination in a tumor considered unresponsive to standard immunotherapy.
Why it matters: It shows personalized neoantigen vaccines are safe and immunogenic in PDAC, a cancer unresponsive to standard immunotherapy.
Tumor-reactive T cells in the bone marrow of patients with multiple myeloma and acute myeloid leukemia carry a conserved effector program distinct from the exhausted phenotype seen in solid cancers, and respond to noncanonical peptides. A transcriptional classifier, TFiT, identifies these cells and stratifies immunotherapy, but not chemotherapy, response across independent MM and AML cohorts. The work defines a latent, activatable anti-tumor T cell compartment in bone marrow malignancies.
Why it matters: It provides a classifier that predicts immunotherapy response in myeloma and AML from T cell state.
PLA2G2D-expressing myeloid cells in tumour-draining lymph nodes form a spatial neighbourhood linked to distant metastases in melanoma, and are far more abundant in TDLNs than in primary tumours or metastases. Blocking PLA2G2D genetically or with antibodies markedly reduced tumour growth, expanded tumour-specific T cells in TDLNs, circulation and tumours, and acted as a non-redundant checkpoint with PD-1, with combination treatment showing additive or synergistic efficacy in humanized mice. It identifies a targetable, lymph-node-centred immune checkpoint distinct from PD-1.
Why it matters: It names a new targetable immune checkpoint in tumour-draining lymph nodes that is non-redundant with PD-1 blockade.
The FDA has approved Bexlutry, Curium's lutetium Lu 177 dotatate injection, as the industry's first radioligand equivalent. The approval for the Boston company follows a court ruling that cleared Curium of all patent claims brought by Novartis, whose reference product is Lutathera.
Why it matters: First FDA-approved radioligand equivalent, a copycat of Novartis' Lutathera, after Novartis lost its patent case.
China's NMPA approved trastuzumab deruxtecan (T-DXd) combined with pertuzumab on August 11, 2026 as first-line treatment for adults with unresectable or metastatic HER2-positive breast cancer. It is the first first-line ADC regimen approved in HER2-positive metastatic breast cancer in China. The approval rests on the global DESTINY-Breast09 study, in which Chinese patients made up close to a quarter of enrolment and the Chinese subgroup showed efficacy and safety better than the global population.
Why it matters: First first-line ADC approval in HER2-positive metastatic breast cancer in China, with strong local subgroup data.
The FDA has approved Telix's Pixclara (floretyrosine F 18, or 18F-FET), the first FET-PET imaging drug for glioma approved in the U.S., more than a year after rejecting the application in April over a need for additional confirmatory clinical evidence. The diagnostic, used with PET to distinguish recurrent or progressive glioma from treatment-related change, is approved for adults and children one month of age and older. Gliomas account for around 80% of malignant brain tumors, with around 24,000 U.S. cases yearly.
Why it matters: First FDA-approved FET-PET imaging drug for glioma fills a guideline-recommended diagnostic gap in the U.S.
CD55, classically a complement regulator, acts in NK cells as a tumor immunity enhancer. Tumor contact induces NK cells to upregulate CD55, which binds CD97 on tumor cells, promotes lipid raft clustering and activates LCK signaling, increasing NK cell killing, cytokine secretion and proliferation. Sustained loss of CD55 in the tumor microenvironment impairs NK cell function, and restoring CD55 expression improves anti-tumor efficacy of NK cells and CAR-NK cells.
Why it matters: Identifies CD55 loss in the tumor microenvironment as a driver of NK cell dysfunction and a target to improve NK and CAR-NK therapy.
STAT3 inhibition eliminates multiple myeloma stem-like cells and elicits hallmarks of immunogenic cell death, Sha et al. report in Blood Science. The findings link stemness, the integrated stress response, and antitumor immunity in multiple myeloma, with in vivo validation still needed before the strategy can be considered immunogenic.
Why it matters: It connects stemness, integrated stress response, and antitumor immunity in a strategy against multiple myeloma.
The oral dual on-off binding KRAS G12D inhibitor GFH375 was evaluated in a phase 1 trial in patients with previously treated KRASG12D-mutated advanced solid tumors. Most treatment-related adverse events were low-grade, but there was one treatment-related death. Response rates were described as encouraging in non-small cell lung cancer and pancreatic cancer.
Why it matters: First-in-human phase 1 data on an oral KRAS G12D inhibitor across KRASG12D-mutated solid tumors.
CIRI3 detects and quantifies circular RNAs in terabyte-scale RNA-seq datasets, running an order of magnitude faster than existing tools with increased accuracy through dynamic multithreaded task partitioning and a blocking search strategy for junction reads. Applied to 2,535 cancer-related samples, it identified differentially spliced circRNAs and yielded the CIRIonco database with a pretraining model and biomarker network.
Why it matters: A much faster and more accurate tool makes large-scale circRNA analysis across cancer datasets practical.
Gotistobart, a next-generation CTLA-4 antibody from BioNTech and OncoC4, nearly doubled median overall survival in previously treated squamous non-small cell lung cancer in updated stage 1 data from the phase 3 PRESERVE-003 trial. Median overall survival was 18.5 months with gotistobart versus 10.0 months with standard-of-care chemotherapy. The data support its mechanism of selectively depleting regulatory T cells in the tumor microenvironment.
Why it matters: Phase 3 data show a near-doubling of median overall survival with gotistobart monotherapy in previously treated squamous NSCLC.
GFH375 monotherapy at 600 mg once daily produced an objective response rate of 59.2% and a confirmed objective response rate of 52.1% in previously treated KRAS G12D-mutant non-small cell lung cancer, with a disease control rate of 93% among 71 evaluable patients. Nearly all enrolled patients had distant metastases and over 40% had received two or more prior lines. The updated phase 2 data were presented in an oral report at the 2026 World Conference on Lung Cancer in Seoul.
Why it matters: Phase 2 efficacy data for a KRAS G12D inhibitor in heavily pretreated NSCLC, selected for oral presentation at WCLC 2026.
The clinically used drug 5-azacytidine kills leukemia cells by depleting 5-methylcytosine (m5C) on chromatin-associated RNA, not only through DNA hypomethylation. Loss of m5C disrupts MBD6-mediated removal of H2AK119ub and SRSF2-mediated deposition of H3K27ac, leading to cell-cycle disruption and defective DNA repair. This reframes the drug's mechanism of action in leukemia.
Why it matters: Reassigns 5-azacytidine's cytotoxicity to RNA m5C depletion, altering how its action in leukemia is understood.
Anal squamous cell carcinoma (ASCC) carries a distinct anaerobic microbial signature that separates it from high-grade squamous intraepithelial lesions (HSIL). In shotgun metagenomic sequencing of anal swabs from 121 adults (102 HSIL, 19 treatment-naive nonmetastatic ASCC), six anaerobic species, including Fusobacterium nucleatum and Parvimonas micra, were enriched in ASCC, persisted across longitudinal sampling, and declined after treatment initiation. The finding points to microbial features that may help distinguish invasive anal cancer from precursor lesions.
Why it matters: It identifies a reproducible microbiome signature that could mark progression from HSIL to invasive anal cancer.
Glioma-derived guanidinoacetate (GAA) depolarizes neighboring neurons through GABAA receptors, and depleting this metabolite reduces both neuronal activity and glioma growth.
Why it matters: The study shows gliomas use the metabolite GAA to influence neuronal activity and promote their own growth, offering a new angle on targeting tumor microenvironment metabolic interactions.
Cross-source progression assessment in advanced pancreatic cancer frequently failed to yield immediate confirmation of progression. Among 645 patients in the primary estimand, only 25 (3.88%) showed same-day concordance, 219 (33.95%) confirmation, 108 (16.74%) explicit nonprogression, and 166 (25.74%) were indeterminate. Among classifiable first opportunities, 33.03% documented explicit nonprogression, supporting an opportunity-adjusted six-state framework over binary discordance estimates.
Why it matters: Quantifies how often oncology and radiology sources disagree on progression, affecting how progression endpoints are interpreted.
