Nature Cancer

Children with relapsed or refractory acute lymphoblastic leukemia (ALL) require more effective and less toxic therapies. We established a prospective, multicenter Drug Response Profiling (DRP) registry (NCT06550102) integrating functional testing into precision-guided treatment. DRP was performed for 340 patients from 17 European countries with a turn-around time of two-weeks. Image-based drug screening with over 135000 unique perturbations revealed a heterogeneous landscape of ex vivo responses to 88 drugs on average. Ranking drug responses across the patient cohort defined individual drug fingerprints, identifying "DRP twins" by similarity in sensitivity and resistance independent of genetic ALL subtypes. Of 239 high-risk patients with follow-up, DRP-informed interventions were reported for 63 patients (26%). Patients received combination therapies based on venetoclax, tyrosine kinase inhibitors, trametinib, bortezomib or selinexor, resulting in objective clinical responses in 43 cases (68%). Precision-guided treatments allowed bridging to cellular therapies in 42 patients among whom 28 (67%) were still alive with a median follow-up of 21 months after DRP (IQR: 14.7-26.6 months). Top responders to venetoclax, ranked within the first tertile of the cohort, had superior 1-year event-survival compared to venetoclax non-responders (0.57 [95% CI, 0.39-0.85] vs. 0.25 [95% CI, 0.11-0.58]). Collectively, these findings demonstrate the feasibility and clinical relevance of functional profiling within an international network. This scalable framework enables individualized therapy selection for enrolment in adaptive precision trials for high-risk pediatric ALL.

Cell-free DNA (cfDNA) profiling enables minimally invasive cancer detection and monitoring. We present SIMMA, a low-input single-molecule sequencing approach that enables multimodal whole-genome and high-depth targeted sequencing of the same cfDNA sample for both tumour-agnostic and tumour-informed liquid biopsy analysis. Across 792 plasma and cerebrospinal fluid cfDNA samples from 277 paediatric patients with diverse brain and extracranial tumours, SIMMA enabled tumour diagnosis, detection of driver mutations, and reconstruction of extrachromosomal DNA (ecDNA) months before clinical relapse. Using conformal prediction trained on genome-wide fragmentomics, genomic and epigenomic data, SIMMA predicts disease burden as a continuous variable and provides well-calibrated uncertainty estimates for each sample, achieving a limit of detection of [~]100 ppm from low-pass whole-genome sequencing data. In summary, SIMMA establishes the clinical utility of multimodal cfDNA profiling with uncertainty quantification for individual patients and unlocks the potential of ecDNA as a liquid biopsy biomarker for disease detection and monitoring across diverse aggressive malignancies.

Background: Predicting whether a treatment will demonstrate meaningful clinical benefit before committing to a large-scale trial remains a major unmet need in oncology. Patient-derived organoids (PDOs) recapitulate individual tumor drug sensitivity, but have not been used to forecast population-level trial outcomes. We developed SCOPE (Screening-to-Clinical Outcome Prediction Engine), a platform that integrates PDO drug screening with clinical prognostic modeling to predict arm-level median progression-free survival (mPFS) and objective response rate (ORR) without access to any trial outcome data. Patients and methods: SCOPE was trained on 54 treatment lines from patients with metastatic colorectal cancer (mCRC, n=15) and metastatic pancreatic ductal adenocarcinoma (mPDAC, n=39) with matched clinical data and PDO drug screening across 9 compounds. A Clinical Score module captures baseline prognosis; a Drug Screen Score module quantifies treatment-specific organoid sensitivity. To predict trial outcomes, synthetic patient profiles are generated from published eligibility criteria and matched to a biobank of 81 PDO lines. Predictions were externally validated against 32 arms from 23 published trials, treatment ranking was assessed across 8 head-to-head comparisons, and prospective applicability was tested for daraxonrasib (RMC-6236), a novel pan-RAS inhibitor in mPDAC. Results: Predicted mPFS strongly agreed with published outcomes (R2=0.85, MAE=0.82 months; Pearson r=0.92, P<0.001), approaching the empirical concordance between two independently measured clinical endpoints (ORR vs. mPFS, R2=0.87). ORR prediction was similarly robust (R2=0.71, MAE=7.3 percentage points). Integrating organoid and clinical data significantly outperformed either alone (P=0.001). SCOPE correctly identified the superior arm in 7 of 8 head-to-head comparisons (88%, P<0.05). Applied to daraxonrasib prior to phase 3 data availability, the platform predicted superiority over standard chemotherapy in KRAS-mutant mPDAC, consistent with emerging clinical data. Conclusion: By combining functional organoid drug screening with clinical modeling, SCOPE generates calibrated efficacy predictions for both established regimens and novel agents without prior clinical data. This approach could support clinical trial design, treatment arm selection, and go/no-go decisions, offering a new tool to improve the efficiency of gastrointestinal cancer drug development.

Children with acute lymphoblastic leukemia (ALL) are treated with multiagent chemotherapy that causes profound changes to the immune system. There are limited data on how disease and therapy impact antigen-specific immune memory, leading to inconsistent guidelines on best practices for revaccination of this population. Here, to inform vaccine guidance, we investigated whether immunity derived from routine childhood measles and varicella zoster virus (VZV) vaccines is maintained during and after therapy for childhood ALL. We report that antibodies against measles and VZV were significantly reduced in children with ALL (n=45) compared to healthy controls (n=13), particularly in older children in whom a longer time had passed since their most recent vaccine dose. However, the avidity of the measles and VZV-specific antibodies was indistinguishable between groups. Despite changes to the composition of the T cell compartment, both overall and antigen-specific T cell function were preserved in children with ALL. These data provide compelling evidence for revaccination of children following ALL treatment. Intact T cell responses suggest that post-treatment revaccination would be effective.

Background: Stomach adenocarcinoma is driven by heterogeneity, limiting therapeutic success. Although ROS acts as a continuous redox rheostat for tumor evolution, it is categorized based on binary models that are masked by tumor-microenvironment (TME) confounders. Here, we have defined a continuous, TME-independent ROS axis to help identify intrinsic vulnerabilities and improve patient stratification. Methods: Non-negative matrix factorization (NMF) defined a ROS-Axis in TCGA-STAD which was validated in ACRG Cohort. Multivariate regression model isolated intrinsic signatures via residual ROS scores by adjusting for TME confounders. Survival was assessed using Cox hazard models. Drug sensitivities were mapped using GDSC2/ElasticNet modeling with cross-cohort replication. Results: Our results define a reproducible ROS gradient, driven by effectors like NQO1 and SOD1, characterizing ROS-high tumors as proliferative, epithelial and immune -cold. High residual ROS score was associated with an improved prognosis, regardless of TNM stage and age. Pharmacogenomic mapping revealed an overlapping sensitivity to mTOR inhibitors in ROS-high gastric cancer tumors which persisted after TME confounder adjustment. Conclusion: The continuous ROS axis provides a functional readout of metabolic dependency that refines traditional anatomical staging. By identifying mTOR dependent cold tumors, our framework offers a precision strategy for immunotherapy-resistant patients like those affected by microsatellite-stable gastric cancer.

Background and aims: Pancreatic ductal adenocarcinoma (PDAC) surveillance is limited to individuals with familial or genetic risk although most future cases arise outside these groups. In a retrospective study, PRISM, an electronic health record (EHR)-based PDAC risk model, identified individuals in the general population at elevated near-term risk of PDAC. We aimed to prospectively evaluate whether PRISM can identify high-risk individuals beyond current surveillance groups across U.S. health systems. Methods: We performed a prospective multicenter cohort study after deployment of PRISM in April 2023 across 44 U.S. health care organizations. Eligible adults aged [&ge;]40 years without prior PDAC received a single baseline risk score and were assigned to prespecified risk tiers. Patients were followed for incident PDAC for 30 months. We estimated tier-specific 30-month cumulative incidence (positive predictive value, PPV), number needed to screen (NNS), standardized incidence ratios (SIRs), and time from deployment and first high-risk flag to diagnosis. Results: Among 6,282,123 adults assigned a PRISM score, 5,058,067 had follow-up; 3,609 developed PDAC. The highest-risk tier had 30-fold higher PDAC incidence than the study population. At the SIR 5 threshold, 30-month cumulative incidence was 0.35% (NNS, 284.2); at SIR 16, 1.14% (NNS, 87.4); and at SIR 30, 2.19% (NNS, 45.7). Median time from deployment to PDAC diagnosis was 9.5 months, and median time from first high-risk flag to diagnosis at SIR 5 was 3.5 years. Shapley additive explanations (SHAP) analyses supported patient- and tier-level interpretability. Conclusions: Prospective deployment of PRISM across multiple U.S. health care organizations identified individuals at elevated near-term risk for PDAC, with substantial risk enrichment and lead time before diagnosis. These findings support the real-world scalability and generalizability of EHRbased risk stratification for risk-adapted early detection. ClinicalTrials.gov identifier NCT05973331

Effective predictive biomarkers for immune checkpoint inhibitor (ICI) therapy remain an unmet need across solid tumors. Here, we present an integrated spatial proteomics workflow that combines in situ proximity ligation assay with multiplexed immunofluorescence to directly resolve PD1/PDL1 signaling events at the level of defined cellular phenotypes and their spatial organization within intact tumor tissue. Applied as a proof of concept to tumor samples from patients with metastatic urothelial carcinoma treated with pembrolizumab, this approach reveals that PD1/PDL1 interactions specifically involving cytotoxic CD8CD3 T cells are significantly enriched in complete responders, while such interactions are rare in patients with progressive disease. This interaction defined T cell subset achieves superior discrimination of clinical response compared to single marker PDL1 expression or immune cell abundance alone. By integrating direct detection of protein protein interactions with high dimensional single cell phenotyping, our workflow provides a mechanistically informed, spatially resolved biomarker of functional immune engagement. Beyond urothelial carcinoma, this platform establishes a generalizable framework for translating spatial signaling biology into predictive tools for immunotherapy response across tumor types.

Background: Sezary syndrome (SS) is an aggressive leukemic variant of cutaneous T-cell lymphoma (CTCL) with distinct clinical and biological features compared to rarer entities such as primary cutaneous CD8+ aggressive epidermotropic cytotoxic T-cell lymphoma (PCAECTCL). Although recurrent genomic alterations in CTCL have been described, comparative analyses at the pathway level across biologically divergent subtypes remain limited. Here, we leveraged a conversational artificial intelligence (AI) platform for precision oncology to enable rapid, integrative, and hypothesis-driven interrogation of publicly available genomic datasets. Methods: We conducted a secondary analysis of somatic mutation and clinical data from the Columbia University CTCL cohort accessed via cBioPortal. Cases were stratified into SS (n=26) and PCAECTCL (n=13). High-confidence coding variants were curated and mapped to biologically relevant signaling pathways and functional gene categories implicated in CTCL pathogenesis. Pathway-level mutation frequencies were compared using Chi-square or Fisher's exact tests, with effect sizes quantified as odds ratios. Tumor mutational burden (TMB) was compared using the Wilcoxon rank-sum test. Subtype-specific co-mutation patterns were evaluated using pairwise association analyses and visualized through oncoplots and network heatmaps. Conversational AI agents, AI-HOPE, were used to iteratively refine cohort definitions, prioritize pathway-level signals, and contextualize findings. Results: TMB was comparable between SS and PCAECTCL (p = 0.96), indicating no significant difference in global mutational load. In contrast, pathway-centric analyses revealed marked qualitative differences. SS demonstrated enrichment of alterations in epigenetic regulators, tumor suppressor and cell-cycle control pathways, NFAT signaling, and DNA damage response mechanisms, consistent with transcriptional dysregulation and immune modulation. PCAECTCL exhibited relatively higher frequencies of alterations involving epigenetic regulators and MAPK pathway signaling, suggesting distinct oncogenic dependencies. Co-mutation analysis revealed a more constrained and focused interaction landscape in SS, whereas PCAECTCL displayed broader and more heterogeneous co-mutation networks, indicative of divergent evolutionary trajectories. Notably, ERBB2 mutations were significantly enriched between subtypes (p = 0.031), highlighting a potential subtype-specific therapeutic vulnerability. Conclusions: This study demonstrates that SS is distinguished from PCAECTCL not by increased mutational burden but by distinct pathway-level architectures, particularly involving epigenetic regulation, immune signaling, and transcriptional control. These findings generate biologically grounded, testable hypotheses for subtype-specific therapeutic targeting and underscore the value of conversational AI as a scalable framework for accelerating discovery in translational cancer genomics.

Many genetically validated targets in cancer, including the transcription factor {beta}-catenin ({beta}-cat), have historically been viewed as undruggable. Cell-based phenotypic screening of chemical compounds can reveal new biological and pharmacological principles. Natural products are powerful probes because of their superior structural diversity, drug-like properties, and biological activities as compared to unoptimized synthetic compounds. We screened 326,304 natural product mixtures (40,744 extracts and 285,560 fractions derived from them) using mammalian cells expressing an oncogenic version of {beta}-cat fused to a suicide protein. Multiple fractions degraded the {beta}-cat fusion protein or drove it into a compartment where both fusion partners were apparently inactive. The active natural product from one of the latter specifically activates novel, but not classical, protein kinase Cs (PKCs) and thereby relocates {beta}-cat to juxtamembrane vacuolar structures. These findings suggest a path for inactivating oncogenic {beta}-cat and underscore the power of screening natural product collections with robust phenotypic assays.

Background: Current deep learning models in computational pathology, radiology, and digital pathology produce opaque predictions that lack the explainable artificial intelligence (xAI) capabilities required for clinical adoption. Despite achieving radiologist-level performance in tasks from whole-slide image (WSI) classification to mammographic screening, these models function as black boxes: clinicians cannot trace predictions to specific biological features, verify outputs against established morphological criteria, or integrate AI reasoning into precision oncology workflows and tumor board decision-making. Methods: We present Virtual Spectral Decomposition (VSD), a modality-agnostic, interpretable-by-design framework that decomposes medical images into six biologically interpretable tissue composition channels using sigmoid threshold functions - the same mathematical structure as CT windowing. Unlike post-hoc xAI methods (Grad-CAM, SHAP, LIME) applied to black-box deep learning models, VSD channels have pre-defined biological meanings derived from tissue physics, providing inherent explainability without sacrificing quantitative rigor. For whole-slide image (WSI) analysis in digital pathology, we introduce the dendritic tile selection algorithm, a biologically-inspired hierarchical architecture achieving 70-80% computational reduction while preferentially sampling the tumor immune microenvironment. VSD is validated across three cancer types and imaging modalities: pancreatic ductal adenocarcinoma (PDAC) on CT imaging, lung adenocarcinoma (LUAD) on H&E-stained pathology slides using TCGA data, and breast cancer on screening mammography. Composition entropy of the six-channel vector is computed as a visual Biological Entropy Index (vBEI) - an imaging biomarker quantifying the diversity of active biological defense systems. Results: In pancreatic cancer, the fat-to-stroma ratio (a novel CT-derived radiomics biomarker) declines from >5.0 (normal) to <0.5 (advanced PDAC), enabling early detection of desmoplastic invasion before mass formation on standard imaging. In lung cancer, composition entropy from H&E whole-slide images correlates with tumor immune microenvironment markers from RNA-seq (CD3: rho=+0.57, p=0.009; CD8: rho=+0.54, p=0.015; PD-1: rho=+0.54, p=0.013) and predicts overall survival (low entropy immune-desert phenotype: 71% mortality vs 29%, p=0.032; n=20 TCGA-LUAD), providing immune phenotyping for checkpoint immunotherapy patient selection from a $5 H&E slide without molecular assays. In breast cancer, each lesion type produces a characteristic six-channel fingerprint functioning as an interpretable computer-aided diagnosis (CAD) system for quantitative BI-RADS assessment and subtype classification (IDC vs ILC vs DCIS vs IBC). A five-level xAI audit trail provides complete traceability from clinical decision support output to specific biological structures visible on the original images. Conclusion: VSD establishes a unified, interpretable-by-design mathematical framework for explainable tissue composition analysis across imaging modalities and cancer types. Unlike black-box deep learning and post-hoc xAI approaches, VSD provides inherently interpretable, clinically verifiable cancer detection and immune phenotyping from standard clinical imaging at existing costs - without requiring foundation model infrastructure, specialized hardware, or molecular assays. The open-source pipeline (Google Colab, Supplementary Material) enables immediate reproducibility and extension to additional cancer types across the pan-cancer TCGA atlas.

Dementia incidence has been declining in Western societies for decades, but whether this reflects higher cognitive capacity entering old age, slower cognitive decline, or both remains unresolved. Analysing ~783,000 episodic memory assessments from ~219,000 individuals across five longitudinal cohorts, we find that later-born cohorts benefit from a double dividend: higher memory levels entering old age and slower rates of decline. The projected 20-year cohort advantage at age 80 is of sufficient magnitude to plausibly account for the observed 13% per-decade decline in dementia incidence reported in meta-analyses. Generational gains are disproportionately concentrated among the fastest-declining individuals, and are reflected in lower hippocampal atrophy rates in an independent sample. A formal bounding analysis shows that the double dividend is robust across a range of plausible period assumptions, consistent with environmental conditions operating across the lifespan having reshaped the architecture of human cognitive aging.

Background. Pancreatic ductal adenocarcinoma (PDAC) has a five-year survival rate of approximately 12%, largely because it is typically diagnosed at an advanced stage. CT-based computational methods for early detection exist but rely on black-box deep learning or large texture feature sets without tissue-specific interpretability. Methods. We developed Virtual Spectral Decomposition (VSD), which applies six parameterized sigmoid functions S(HU) = 1/(1+exp(-alpha x (HU - mu))) to standard portal-venous CT, decomposing each pixel into tissue-specific response channels for fat (mu=-60), fluid (mu=10), parenchyma (mu=45), stroma (mu=75), vascular (mu=130), and calcification (mu=250). Dendritic Binary Gating identifies structural content per channel using morphological filtering, enabling co-firing analysis and lone firer identification. A 25-feature signature was extracted per patient. Three independent datasets were analyzed: NIH Pancreas-CT (n=78 healthy), Medical Segmentation Decathlon Task07 (n=281 PDAC, paired tumor/adjacent tissue), and CPTAC-PDA from The Cancer Imaging Archive (n=82, multi-institutional, with DICOM time point tags). The same six sigmoid parameters were used across all datasets without retraining. Results. VSD achieved AUC 0.943 for field effect detection (healthy vs cancer-adjacent parenchyma) and AUC 0.931 for patient-stratified tumor specification on MSD. On CPTAC-PDA, VSD achieved AUC 0.961 (6 features) and 0.979 (25 features) for distinguishing healthy from cancer-bearing pancreas on scans obtained prior to pathological diagnosis. All significant features replicated across datasets in the same direction: z_fat (d=-2.10, p=3.5e-27), z_fluid (d=-2.76, p=2.4e-38), fire_fat (d=+2.18, p=1.2e-28). Critically, VSD severity did not correlate with days-from-diagnosis (r=-0.008, p=0.944) across a range of day -1394 to day +249. Patient C3N-01375, scanned 3.8 years before pathological diagnosis, had VSD severity 1.87, well above the healthy mean of 0.94 +/- 0.33. The tissue transformation signature was temporally stable, indicating an early, persistent tissue state rather than a progressively worsening process. Conclusions. VSD with Dendritic Binary Gating detects a stable pancreatic tissue composition signature on standard CT that is present years before clinical diagnosis, validated across three independent datasets without parameter adjustment. The six sigmoid channels map to biologically meaningful tissue components through a fully transparent interpretability chain. The temporal stability of the signal implies a detection window of 3-7 years, consistent with known PanIN-3 microenvironment transformation timelines. VSD functions as a single-scan screening tool applicable to any abdominal CT performed during the pre-clinical window.

Chimeric antigen receptor T-cell (CAR-T) therapy targeting CD19 has transformed outcomes for children with relapsed or refractory (R/R) B-cell acute lymphoblastic leukemia (B-ALL), yet the influence of molecular subtype on outcomes remains unclear. We evaluated the impact of cytogenetic and molecular signatures on complete response (CR), overall survival (OS), and leukemia-free survival (LFS) after CD19 CAR-T therapy in eighty-six pediatric patients with R/R B-ALL treated with tisagenlecleucel. CR was assessed 30 days after infusion. Cytogenetic data were available for 84 patients and molecular profiling for 62. Survival analyses included 72 patients who received CD19 CAR-T as the sole cellular therapy. Seventy-seven patients achieved CR (89.5%). Pre-infusion bone marrow blasts of [&ge;]20% were associated with lower CR rates (53.8% vs 95.9%, p<0.0001) and significantly reduced OS and LFS (both p<0.0001). Among molecular markers, RAS mutations correlated with inferior OS (p=0.0222) and LFS (0.0402). In multivariate analysis, bone marrow blasts >20% and RAS mutations independently predicted inferior OS. Post CAR-T, CD19 negative relapses showed almost twice higher prevalence of RAS mutations (66% vs 37.5%). These findings highlight RAS mutations as a key molecular predictor of outcome after CD19 CAR-T therapy and suggest emergence of unique risk stratification for patients receiving CD19-targeting therapy.

The transcription coregulator OCA-B promotes CD4+ T cell memory recall responses and autoimmunity. OCA-B T cell deletion prevents spontaneous type-1 diabetes (T1D) onset in non-obese diabetic (NOD) mice and blunts T1D in a subset of more aggressive models. However, the role of OCA-B in diabetes induced by treatment with immune checkpoint inhibitors (ICIs), and the role of OCA-B in the control of tumors with and without ICI treatment, has not been studied. Here we show that islet and pancreatic lymph node T cells from T1D individuals express measurable POU2AF1 mRNA. Deletion of OCA-B in T cells fully insulates 8-week-old non-obese diabetic (NOD) mice against ICI-induced diabetes and partially protects 12-week-old mice. Salivary and lacrimal gland infiltration and inflammation were also reduced. Protection was associated with a block in the differentiation of progenitor exhausted CD8+ T cells (TPEX) into terminally exhausted CD8+ T cells (TEX). We show that OCA-B T cell loss preserves anti-tumor immune responses following PD-1 blockade in different tumors and mouse strains. These findings point to a potential therapeutic window in which pharmaceuticals targeting OCA-B could be used to block the emergence of both spontaneous and ICI-induced autoimmunity while sparing anti-tumor immunity. We develop first-in-class small molecule inhibitors of Oct1/OCA-B transcription complexes and show that administration into NOD mice also blocks diabetes emergence following PD-1 blockade. These results identify OCA-B as a promising therapeutic target for the prevention of autoimmunity and immune-related adverse events (irAEs).

BackgroundTreatment optimization in HPV-associated oropharyngeal cancer (OPSCC) remains challenging, as recent de-escalation trials have shown limited success. Current patient selection strategies based on smoking history and TNM classification are insufficient, highlighting the need for robust, standardized prognostic biomarkers. We report the first validation of the Immunoscore (IS) for prognostic stratification in HPV-associated OPSCC. Patients and methodsWe analyzed 191 HPV-associated (p16+ and HPV DNA/RNA+) OPSCC patients from an international multicenter cohort (2015-2024), comprising a French monocentric retrospective training cohort (N = 48) and three validation cohorts: French monocentric retrospective (N = 48), French multicenter prospective (N = 50), and US multicenter retrospective (N = 45). IS is a standardized digital pathology assay quantifying CD3lJ and CD8lJ densities in tumor cores and invasive margins, with cut-offs defined in the training cohort and validated across cohorts. Associations with disease-free survival (DFS), time to recurrence (TTR) and overall survival (OS) were assessed, alongside 3RNA-seq and sequential immunofluorescence profiling of immune composition. ResultsMedian age 65; 80% male; 74% smokers; 66% T1-2; 82% N0-1 (AJCC8th). IS-High patients demonstrated superior 3-year DFS in the training and validation cohorts 1-3 (all log-rank P < 0.05). Multivariable analysis identified IS-Low as the strongest independent risk factor for DFS (HR 9.03; 95% CI: 4.02-20.31; P < 0.001). The model combining IS with clinical factors showed higher predictive accuracy for DFS (C-index 0.82) than clinical variables alone (0.7; P < 0.0001). Similar findings were observed for TTR and OS. IS-High tumors showed markedly higher enrichment of lymphoid and myeloid immune cell populations, contrasting with immune-poor signatures in IS-Low tumors. ConclusionsIS is a robust biomarker that outperforms standard clinical variables in both prognostic and predictive accuracy. The enriched cytotoxic immune infiltrate in IS-High tumors explains favorable outcomes and supports their suitability for treatment de-escalation. Prospective validation is warranted.

BackgroundPlasma biomarkers demonstrate strong within-cohort performance for identifying cerebral amyloid pathology, but their real-world clinical utility depends on generalization across populations and assay platforms. The impact of cross-cohort deployment on clinically actionable metrics such as negative predictive value (NPV) remains poorly characterized. ObjectiveTo evaluate the performance and portability of plasma biomarker-based machine learning models for amyloid PET prediction across independent cohorts, with emphasis on calibration and clinically relevant predictive values. MethodsData from ADNI (n=885) and A4 (n=822) were analyzed. Machine learning models were trained within each cohort to predict amyloid PET status and continuous amyloid burden (centiloids). Performance was assessed using ROC AUC, accuracy, R{superscript 2}, and RMSE. Cross-cohort generalizability was evaluated using bidirectional transfer without retraining. Calibration, predictive values, and decision curve analysis were used to assess clinical utility. ResultsWithin-cohort discrimination was high (AUC up to 0.913 in ADNI and 0.870 in A4), with moderate performance for centiloid prediction (R{superscript 2} up to 0.628 and 0.535, respectively). Cross-cohort deployment resulted in modest attenuation of AUC ([~]4-7%) but substantially greater degradation in clinically actionable performance. NPV declined from 0.831 to 0.644 under ADNI[-&gt;]A4 transfer ([~]19 percentage points) despite preserved discrimination. Calibration analyses demonstrated systematic probability misestimation, and decision curve analysis showed reduced net clinical benefit. Biomarker distribution differences across cohorts were consistent with dataset shift. ConclusionPlasma biomarker models retain discrimination across cohorts but exhibit clinically meaningful degradation in predictive value under deployment. Calibration instability and prevalence differences critically affect NPV, highlighting the need for cross-cohort validation, calibration assessment, and assay harmonization before clinical implementation.

Cancer-related casualties are the most common cause of death worldwide. The discovery of biomarkers is of utmost importance for diagnosis and disease monitoring. Herein, we performed a comprehensive metabolomics biomarker discovery effort in plasma from 615 lung, ovarian and colorectal cancer patients at diagnosis and 95 non-cancerous control subjects. This pan-cancer investigation identified specific panels of metabolites in the entire sample cohort with a high discriminating power and demonstrated by combined ROC AUC values of up to 0.95. The identified metabolites are mainly associated with lipid and amino acid metabolism as well as xenobiotic transformation. These metabolite panels of high predictive power provide new metabolic insights in these cancers and demonstrate the potential of metabolomics for improved diagnosis and monitoring disease progression.

Lung cancer screening saves lives, yet uptake remains suboptimal and inequitable. Personalised communication can improve attendance and reduce anxiety, but scaling such support is a workforce challenge. We fine-tuned Googles Gemma 2 9B using QLoRA on 5,086 synthetic screening conversations and compared it against Googles Gemini 2.5 Flash (a larger frontier model) and an unmodified baseline across 300 multi-turn conversations with 100 patient personas spanning ten clinical categories. Evaluation combined automated natural language processing metrics with independent language model judgement in two complementary modes: structured clinical rubric and simulated patient persona. The fine-tuned model achieved the highest simulated patient experience score (3.71/5 vs 3.65 for the frontier model), recorded zero boundary violations after clinician review of all flagged instances, and led on the four most safety-critical categories. A composite Patient Adaptation Index showed that the fine-tuned model led overall (0.37 vs 0.35 vs 0.35), with its clearest advantage on the two clinically specific components: empathy calibration to patient distress and selective smoking cessation signposting. These findings suggest that targeted fine-tuning of open-source models can yield clinical communication quality comparable to larger proprietary systems, with advantages in safety-critical scenarios and suitability for NHS data governance constraints. Human clinician review of these conversations is ongoing.

Background. Detection of cerebral amyloid pathology currently requires amyloid PET imaging ($5,000-$8,000) or cerebrospinal fluid analysis via lumbar puncture, procedures that are inaccessible for population-level screening. The FDA-cleared Lumipulse G pTau217/Abeta1-42 plasma ratio test (May 2025) represents the first approved blood-based alternative; however, single-ratio approaches cannot distinguish Alzheimer's disease (AD) from non-AD neurodegeneration or provide multi-dimensional disease characterization. Methods. We developed Virtual Spectral Decomposition (VSD), a framework that decomposes plasma biomarker profiles into biologically interpretable diagnostic channels. Four plasma biomarkers - phosphorylated tau-217 (pTau217), amyloid-beta42/40 ratio, neurofilament light chain (NfL), and glial fibrillary acidic protein (GFAP) - were measured in 1,139 Alzheimer's Disease Neuroimaging Initiative (ADNI) participants. Each biomarker was mapped to a VSD channel representing a distinct pathophysiological axis: tau/amyloid phosphorylation, amyloid clearance, neurodegeneration, and astrocytic activation. Channel weights were calibrated via logistic regression, and performance was evaluated against amyloid PET (UC Berkeley) using 10x5-fold repeated cross-validation. Results. VSD 4-channel fusion achieved AUC = 0.900 (+/-0.018), exceeding pTau217 alone (0.888+/-0.022). Optimal sensitivity was 89.7% with 78.1% specificity (NPV = 90.8%). The NfL channel received a negative weight (beta = -1.1), functioning as a disease-exclusion signal: elevated neurodegeneration without amyloid-tau coupling actively reduces the AD probability, distinguishing AD from non-AD neurodegeneration. Complementary CSF proteomics analysis (7,008 proteins, 533 participants) identified 17 amyloid-specific proteins (0.24% of the proteome), revealing a 49:1 tau-to-amyloid asymmetry that explains why blood-based tau markers outperform amyloid markers. Conclusions. Blood-based VSD provides an interpretable, multi-channel framework for amyloid detection that incorporates explicit disease-exclusion logic unavailable to single-biomarker approaches. The architecture extends to multi-disease screening, where the same blood specimen could be routed through disease-specific modules for AD, Parkinson's disease, and cancer.

Amyotrophic lateral sclerosis (ALS) is a rapidly progressing neurodegenerative disease with a heterogeneous clinical presentation, complicating early diagnosis and therapeutic monitoring. To identify disease-specific biomarkers, we performed an unbiased cerebrospinal fluid (CSF) proteomic analysis in 87 ALS patients, 89 healthy controls, and 61 neurological controls using mass spectrometry. Across all quantified proteins, 399 were significantly dysregulated in ALS, including established neurodegeneration (NEFL, NEFM, UCHL1) and neuroinflammatory (CHIT1, CHI3L1, CHI3L2) markers. Correlation and pathway analyses uncovered dysregulation of immune, synaptic, and metabolic processes, with aberrant complement activation emerging as a hallmark. Complement proteins increased progressively with declining ALS Functional Rating Scale-Revised and longer disease duration, whereas early-stage markers (CLSTN3, CHAD, RELN) indicated pre symptomatic neuronal and synaptic disruptions. Machine learning identified a minimal five protein CSF panel (MB, ITLN1, YWHAG, FCGR3A, PGAM1) that accurately distinguished ALS patients from healthy controls, capturing disease-specific pathophysiology beyond general neurodegeneration. Our findings define a robust ALS-specific CSF proteomic signature, reveal prognostic protein candidates across disease stages, and provide a framework for diagnostic biomarker development, enabling earlier intervention and monitoring.