An expanded DNA methylation-based sarcoma classifier (v13.1) was trained on 4377 methylation profiles covering 116 tumor sub-classes, 4 control groups and 93 distinct methylation classes, and validated in 1547 mesenchymal tumors from four independent cohorts. Five-fold nested cross-validation yielded 98% class-level accuracy with a Brier score of 0.017. Predictions were obtained in 73% of validation cases, 91% matched the histopathology diagnosis, and 6% of inconclusive or ambiguous cases gained a definitive molecular diagnosis or reclassification.
Type 2 deiodinase (DIO2), the enzyme converting T4 to T3, is a conserved marker of cancer-associated fibroblasts (CAF) across thyroid cancer and six additional solid tumor types, and is enriched in myofibroblastic CAF. Elevated DIO2 expression correlated with worse overall survival in pancreatic and bladder cancers, while thyroid cancer showed an inverse association linked to loss of epithelial DIO2 during dedifferentiation. In a TPO-CreERT2/BrafV600E/Trp53-deficient mouse model, Dio2 loss in the tumor microenvironment significantly reduced tumor growth, and D2 activity was suppressed by cefuroxime and lenvatinib.
A modeling analysis of the NHS-Galleri trial projects that the reduction in stage IV diagnoses seen after two annual multi-cancer early detection screening rounds could yield 66.3 potential five-year life-years in the trial intervention cohort, or 466 per million people screened annually (95% interval 190-743). Intervention and control arms recorded 180 versus 239 stage IV diagnoses across ten evaluable cancers, with colorectal, lung and esophageal cancers accounting for 79% of the net projection. The authors stress this conditional estimate is not evidence of mortality reduction, and longer follow-up is needed.
Why it matters: Translates NHS-Galleri's stage IV shift into projected life-years per million screened, clarifying what the screening signal may or may not mean.
Evo2, a genomic foundation model, was used to embed short variants, fusions and rearrangements from 102 patients with soft tissue sarcoma and 129 TCGA-SARC cases into patient-level genomic representations for Cox survival modelling. Adding Evo2 to SARCULATOR raised the OS C-index from 0.620 to 0.770 in the study cohort and from 0.595 to 0.716 in TCGA-SARC; in TCGA-SARC it also improved CINSARC-based OS prediction from 0.476 to 0.653. These descriptive point estimates support altered tumor sequence context as an independent prognostic layer for sarcoma risk stratification.
RP-3467, a selective small-molecule inhibitor of DNA polymerase theta, suppresses Polθ enzymatic activity, abrogates microhomology-mediated end joining, and induces synthetic lethality in homologous recombination-deficient models. Combined with PARP inhibitors olaparib or rucaparib, it produced sustained tumor regression across multiple xenograft and patient-derived xenograft models without added systemic toxicity. In a phase 1 trial (NCT06560632), a heavily pretreated germline BRCA1-mutant ovarian cancer patient with a TP53BP1 loss-of-function mutation achieved a confirmed partial response to RP-3467 plus olaparib by RECIST v1.1.
Ataraxis Breast CTX, a causal multi-modal AI model combining clinical variables with H&E histopathology, stratified residual recurrence risk in 556 node-positive HR+/HER2- breast cancer patients from the UNIRAD trial. Five-year disease-free survival was 93% in the low-risk group versus 80% in the high-risk group, and CTX-benefit predicted everolimus benefit in an exploratory multivariable analysis (interaction p = 0.01). The model may inform selection of patients for adjuvant treatment escalation.
BMI at diagnosis shows a U-shaped association with recurrence-free survival in early-onset colorectal cancer, with the lowest hazard at BMI 27 and the highest at BMI below 20 or above 30 (LRT p=0.02). In 1,459 curative-intent surgery patients, no such association appeared in average-onset colorectal cancer (LRT p=0.15), and the cohorts differed significantly (LRT p=0.046). The findings suggest the obesity paradox may be specific to EOCRC.
Melanomas harbor decidual-like NK cells with a CD56 Bright CD16 - CD49a + CD9 + phenotype, a population rare or absent in peripheral blood from patients with melanoma and healthy donors. The cells appear in tumors across sex, melanoma subtype, and disease site, and tumor CD9 expression correlated with intratumoral dl-NK frequency (Pearson rho = 0.40, p = 0.056). Co-culture showed direct contact with CD9-expressing melanoma cells induces CD9 acquisition by NK cells, suggesting an immunologic parallel between melanoma and the maternal-fetal interface.
A methylation-based ctDNA assay (Guardant360) detected recurrence in non-virally associated head and neck squamous cell carcinoma with 76.9% sensitivity and 94.6% specificity (AUC 0.858) in a 50-patient post-treatment cohort. Positive ctDNA preceded clinical detection in 6 of 10 recurrences by a median of 131 days and tracked with worse survival: 1-year progression-free survival was 25.0% versus 96.4%. The findings support adding methylation-based ctDNA to surveillance for this population, pending prospective validation.
An AI model integrating H&E pathology image features with clinical variables (Ataraxis Breast CTX) identified which intermediate-RS node-negative HR+/HER2- breast cancer patients benefit from adjuvant chemotherapy in a prespecified analysis of 6735 TAILORx patients. CTX-benefit showed a significant treatment-by-biomarker interaction (p = 0.016), with high-benefit patients gaining 5.6% in 5-year disease-free interval from chemotherapy while low-benefit patients showed none; stratifying by RS alone at a matching threshold was not significant (p = 0.20). CTX-prognostic also predicted DFI independent of RS (C-index 0.736 with endocrine therapy alone).
A large language model-based resource, MSK-Tox, captured the incidence, timing, and grade of anti-cancer therapy toxicity across more than 50,000 patients. Analysis of six representative adverse events showed toxicity risk varies by cancer type, regimen, and clinical context, is predictable from pretreatment clinical features, and has two germline modes: an organ-intrinsic mode, exemplified by a regulatory variant near FOXE1 linked to hypothyroidism, and an immune-mediated mode in which HLA-DRB1*15 drives adrenal insufficiency after immune checkpoint inhibition. The work supports pretreatment risk assessment for harm on the same principles used to predict benefit.
Distinct sets of chromatin regulator (CR) gene mutations predict improved survival after immune checkpoint inhibitor (ICI) treatment in each of ten cancer types, with some CRs recurrently favorable across cancers. Co-occurrence of mutations in two or more CRs further improved survival, and CR-altered tumors consistently showed higher tumor mutational burden than CR-unaltered tumors. Transcriptomic deconvolution linked CR mutations to cancer-type-specific differences in the tumor microenvironment.
A tractography-derived measure of brain-wide tumor involvement, the lesion–tract density index (L-TDI), outperformed conventional tumor volume for predicting overall survival in 999 patients with newly diagnosed glioblastoma across multiple institutions. Tumor volume's prognostic value depended strongly on analytical design and was largely confined to the contrast-enhancing compartment, whereas L-TDI showed robust associations with overall survival across thresholds and tumor compartments and retained its value after adjustment for established clinical variables. The findings support assessing glioblastoma prognosis as a brain-wide disease rather than by local tumor extent.
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.
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.
Peripheral-blood cfDNA matched or exceeded marrow-plasma cfDNA as a liquid biopsy for multiple myeloma, despite the tumor's marrow confinement. In 74 patients profiled across 372 cfDNA time points, blood cfDNA recovered marrow-defined CNAs and mutations at 77.4% vs 72.0% and 69.2% vs 65.4%, and at high tumor fraction adverse-lesion sensitivity reached 100% in both compartments. Marrow cfDNA did show distinct, handling-sensitive fragmentomic architecture (LOOCV AUC=0.910), so compartment, biomarker class, tumor fraction and processing all shape liquid-biopsy performance.
Why it matters: It shows a simpler blood draw is not inferior to marrow-derived cfDNA for myeloma genomic profiling, with practical processing caveats.
The Sialyl-Thomsen-nouveau (STn) glycovariant of CA-125 predicted ovarian cancer years before clinical diagnosis, whereas conventional CA-125, HE4 and MGL-detected glycovariants did not. In a case-control study of 82 incident ovarian cancer cases matched 1:3 to controls from the Malmö Diet and Cancer cohort, CA-125-STn had an AUC of 0.68 for cases within 5 years of blood draw, and 49% of those cases fell in the top 20% of CA-125-STn levels. The finding suggests a marker that may detect ovarian cancer earlier than conventional CA-125 screening, which two randomized trials showed does not reduce mortality.
A machine-learning classifier detected early-stage ovarian cancer from untargeted metabolomic profiles with cross-validated sensitivity and specificity of 99.0% and 99.8% in serum (59 cancer, 32 controls) and 97.7% and 99.7% in plasma (63 cancer, 20 controls). Both cohorts shared 239 concordantly altered features across lipid, amino-acid, steroid, central-carbon, and redox metabolism, supporting a distributed metabolic 'Systemotype' detectable in blood.
Patient-specific proliferation and diffusion rates of a reaction-diffusion model can be inferred from a single routine H&E biopsy by fitting cell-nucleus point patterns, validated across eleven multicenter cohorts. The derived dispersion length and front velocity stratified progression-free and overall survival independent of routine covariates. A virtual glioblastoma trial allocating hypofractionation to faster-growing tumors and dose intensification to slower-growing ones added 96.7 days of restricted mean progression-free survival over standard of care (permutation p < 0.001).
Why it matters: It turns a single routine H&E biopsy into patient-specific tumor growth parameters usable for survival prediction and virtual treatment trials.
An integrated model combining STAR-CAP cancer-specific risk, OCCAM other-cause mortality risk, and MARCAP efficacy estimates shows wide variation in absolute benefit from androgen deprivation therapy within NCCN risk groups. Applied to 5468 PLCO prostate cancer patients, adding short-term ADT to radiotherapy in unfavorable intermediate-risk disease gave a median 10-year distant metastasis ARR of 4% (range <1% to almost 15%), while prolonging ADT in high-risk disease gave a median ARR of 9% (0.4%-17.2%). The model provides individualized estimates to support treatment discussions rather than uniform group-level recommendations.
