❌

Normal view

Liquid Biopsy in Non-Metastatic Prostate Cancer: Clinical Evidence and Future Directions

Cancers (Basel). 2026 Feb 28;18(5):800. doi: 10.3390/cancers18050800.

ABSTRACT

BACKGROUND AND OBJECTIVE: Liquid biopsy has transformed the management of advanced prostate cancer, yet its clinical role in non-metastatic disease remains uncertain. Conventional biomarkers such as PSA, imaging, and pathology have limited ability to capture minimal residual disease and biological aggressiveness. The objective of this review was to critically evaluate the current evidence on circulating tumor cells (CTCs) and circulating tumor DNA (ctDNA) in non-metastatic prostate cancer, focusing on feasibility, prognostic value, and potential clinical applications.

METHODS: A narrative review of PubMed-indexed original studies evaluating liquid biopsy in clinically localized or non-metastatic prostate cancer was performed. Eligible studies included patients treated with curative-intent local therapy or experiencing biochemical recurrence without radiologic metastases. Study designs were predominantly prospective or retrospective observational cohorts. Liquid biopsy analytes included CTCs and ctDNA assessed from peripheral blood plasma using EpCAM-based enrichment, targeted next-generation sequencing, whole-genome sequencing, or ultra-sensitive tumor-informed assays. Primary outcomes included detection rates, associations with clinicopathologic features, biochemical recurrence, metastasis-free survival, and overall survival. Key Findings and Limitations: Across 11 studies, CTC detection using EpCAM-based platforms was infrequent in localized disease and biochemical recurrence and showed limited prognostic value (10-11% in preoperative settings). In contrast, ctDNA was detectable in a minority of patients but consistently identified biologically aggressive disease and a higher risk of recurrence when present, particularly using tumor-informed ultra-sensitive assays. Limitations include low detection rates, heterogeneous methodologies, small sample sizes, and predominantly exploratory study designs.

CONCLUSIONS AND CLINICAL IMPLICATIONS: Currently, its most promising application is not broad screening, but as a selective, biology-driven tool for detecting minimal residual disease and refining risk assessment. CtDNA acts as a biological risk modifier, potentially guiding the escalation or de-escalation of adjuvant therapy. However, prospective biomarker-driven trials are required to validate these strategies before routine clinical implementation.

PMID:41827734 | PMC:PMC12984391 | DOI:10.3390/cancers18050800

Orally Administered Porcine Intestinal Lactobacilli Improve the Respiratory Innate Immune Response Against <em>Streptococcus pneumoniae</em>

Animals (Basel). 2026 Mar 6;16(5):825. doi: 10.3390/ani16050825.

ABSTRACT

BACKGROUND: Respiratory bacterial infections represent a major health challenge in swine production, highlighting the need for novel immunomodulatory strategies that enhance host resistance. In this study, we investigated whether porcine intestinal lactobacilli could modulate the gut-lung axis and improve respiratory innate immunity in a mouse model of Streptococcus pneumoniae infection, as a surrogate of Streptococcus suis pneumonia.

METHODS: Three strains of Ligilactobacillus salivarius (LAFF998, LAFF1071, and LAFF1095) were orally administered to Swiss mice prior to pneumococcal challenge. The resistance to the infection, the lung damage and the respiratory innate immune response were evaluated.

RESULTS: Only strain LAFF998 significantly reduced pulmonary bacterial loads, prevented bacteremia, and attenuated lung injury. This protective effect was associated with selective modulation of respiratory immunity, characterized by reduced neutrophilic inflammation, increased lymphocyte recruitment, and enhanced activation of alveolar macrophages expressing MHC-II. LAFF998 markedly increased the production of IFN-β, IFN-γ, IL-6, IL-10, and IL-27 in the respiratory tract, without inducing excessive inflammatory damage. Ex vivo and in vitro analyses confirmed that alveolar macrophages from LAFF998-treated mice exhibited a primed phenotype with heightened cytokine responses to pneumococcal stimulation. In contrast, strains LAFF1071 and LAFF1095 failed to confer protection or significantly modulate respiratory immune responses.

CONCLUSIONS: These findings demonstrate a strict strain-dependent effect among porcine L. salivarius isolates and identify LAFF998 as a potent immunobiotic capable of enhancing respiratory innate immunity through the gut-lung axis. This work supports further studies of LAFF998 as an immunobiotic strategy for the prevention of respiratory infections in pigs.

PMID:41829035 | PMC:PMC12985233 | DOI:10.3390/ani16050825

A structure-based mRNA vaccine for Nipah virus in healthy adults: a phase 1 trial

Nature Medicine, Published online: 12 March 2026; doi:10.1038/s41591-026-04265-1

In this phase 1, open-label dose-escalation study in healthy adults found that the mRNA vaccine (mRNA-1215), encoding the Nipah virus Malaysian strain chimeric pre-fusion F protein linked to glycoprotein G, was safe and induced elevated immune responses at 1 year of follow-up, indicating that this is a promising vaccine candidate for further development.

A cognitive layer architecture to support large-language model performance in psychotherapy interactions

Nature Medicine, Published online: 12 March 2026; doi:10.1038/s41591-026-04278-w

A real-world study showed that introducing a cognitive layer architecture to support specialized psychotherapeutic reasoning capabilities in general-purpose chatbots improved depression and anxiety symptoms compared to chatbots or therapists alone.

A sorghum pangenome reference improves global crop trait discovery

Nature, Published online: 11 March 2026; doi:10.1038/s41586-026-10229-9

A pangenome reference for the phenotypically diverse crop sorghum aims to help accelerate future efforts to breed crops that are better adapted to changing environments.

Blood phosphorylated tau elevation as a biomarker in immunoglobulin light chain and transthyretin amyloidosis

Nature Medicine, Published online: 11 March 2026; doi:10.1038/s41591-026-04272-2

Elevated serum levels of phosphorylated tau are not specific to Alzheimer’s disease and may also serve as a diagnostic tool for the most common types of systemic amyloidosis, with potential utility in distinguishing amyloidosis-related polyneuropathy from polyneuropathy of other etiologies.

Ageing promotes metastasis via activation of the integrated stress response

Nature, Published online: 11 March 2026; doi:10.1038/s41586-026-10216-0

Ageing reprograms the evolutionary trajectory of KRAS-driven lung adenocarcinoma, limiting primary tumour growth while promoting metastatic dissemination through epigenetic activation of the integrated stress response, and a therapeutic opportunity in older patients is revealed.

Scaling Laws in the Tiny Regime: How Small Models Change Their Mistakes

arXiv:2603.07365v1 Announce Type: cross Abstract: Neural scaling laws describe how model performance improves as a power law with size, but existing work focuses on models above 100M parameters. The sub-20M regime -- where TinyML and edge AI operate -- remains unexamined. We train 90 models (22K--19.8M parameters) across two architectures (plain ConvNet, MobileNetV2) on CIFAR-100, varying width while holding depth and training fixed. Both follow approximate power laws in error rate: $\alpha = 0.156 \pm 0.002$ (ScaleCNN) and $\alpha = 0.106 \pm 0.001$ (MobileNetV2) across five seeds. Since prior work fit cross-entropy loss rather than error rate, direct exponent comparison is approximate; with that caveat, these are 1.4--2x steeper than $\alpha \approx 0.076$ for large language models. The power law does not hold uniformly: local exponents decay with scale, and MobileNetV2 saturates at 19.8M parameters ($\alpha_{\mathrm{local}} = 0.006$). Error structure also changes. Jaccard overlap between error sets of the smallest and largest ScaleCNN is only 0.35 (25 seed pairs, $\pm 0.004$) -- compression changes which inputs are misclassified, not merely how many. Small models concentrate capacity on easy classes (Gini: 0.26 at 22K vs. 0.09 at 4.7M) while abandoning the hardest (bottom-5 accuracy: 10% vs. 53%). Counter to expectation, the smallest models are best calibrated (ECE = 0.013 vs. peak 0.110 at mid-size). Aggregate accuracy is therefore misleading for edge deployment; validation must happen at the target model size.

Graph-Instructed Neural Networks for parametric problems with varying boundary conditions

arXiv:2603.08304v1 Announce Type: cross Abstract: This work addresses the accurate and efficient simulation of physical phenomena governed by parametric Partial Differential Equations (PDEs) characterized by varying boundary conditions, where parametric instances modify not only the physics of the problem but also the imposition of boundary constraints on the computational domain. In such scenarios, classical Galerkin projection-based reduced order techniques encounter a fundamental bottleneck. Parametric boundaries typically necessitate a re-formulation of the discrete problem for each new configuration, and often, these approaches are unsuitable for real-time applications. To overcome these limitations, we propose a novel methodology based on Graph-Instructed Neural Networks (GINNs). The GINN framework effectively learns the mapping between the parametric description of the computational domain and the corresponding PDE solution. Our results demonstrate that the proposed GINN-based models, can efficiently represent highly complex parametric PDEs, serving as a robust and scalable asset for several applied-oriented settings when compared with fully connected architectures.

Retrieval-Augmented Anatomical Guidance for Text-to-CT Generation

arXiv:2603.08305v1 Announce Type: cross Abstract: Text-conditioned generative models for volumetric medical imaging provide semantic control but lack explicit anatomical guidance, often resulting in outputs that are spatially ambiguous or anatomically inconsistent. In contrast, structure-driven methods ensure strong anatomical consistency but typically assume access to ground-truth annotations, which are unavailable when the target image is to be synthesized. We propose a retrieval-augmented approach for Text-to-CT generation that integrates semantic and anatomical information under a realistic inference setting. Given a radiology report, our method retrieves a semantically related clinical case using a 3D vision-language encoder and leverages its associated anatomical annotation as a structural proxy. This proxy is injected into a text-conditioned latent diffusion model via a ControlNet branch, providing coarse anatomical guidance while maintaining semantic flexibility. Experiments on the CT-RATE dataset show that retrieval-augmented generation improves image fidelity and clinical consistency compared to text-only baselines, while additionally enabling explicit spatial controllability, a capability inherently absent in such approaches. Further analysis highlights the importance of retrieval quality, with semantically aligned proxies yielding consistent gains across all evaluation axes. This work introduces a principled and scalable mechanism to bridge semantic conditioning and anatomical plausibility in volumetric medical image synthesis. Code will be released.
  • ✇cs.AI, q-bio.NC updates on arXiv.org
  • Decomposing Physician Disagreement in HealthBench Satya Borgohain · Roy Mariathas
    arXiv:2602.22758v2 Announce Type: replace Abstract: We decompose physician disagreement in the HealthBench medical AI evaluation dataset to understand where variance resides and what observable features can explain it. Rubric identity accounts for 15.8% of met/not-met label variance but only 3.6-6.9% of disagreement variance; physician identity accounts for just 2.4%. The dominant 81.8% case-level residual is not reduced by HealthBench's metadata labels (z = -0.22, p = 0.83), normative rubric l
     

Decomposing Physician Disagreement in HealthBench

arXiv:2602.22758v2 Announce Type: replace Abstract: We decompose physician disagreement in the HealthBench medical AI evaluation dataset to understand where variance resides and what observable features can explain it. Rubric identity accounts for 15.8% of met/not-met label variance but only 3.6-6.9% of disagreement variance; physician identity accounts for just 2.4%. The dominant 81.8% case-level residual is not reduced by HealthBench's metadata labels (z = -0.22, p = 0.83), normative rubric language (pseudo R^2 = 1.2%), medical specialty (0/300 Tukey pairs significant), surface-feature triage (AUC = 0.58), or embeddings (AUC = 0.485). Disagreement follows an inverted-U with completion quality (AUC = 0.689), confirming physicians agree on clearly good or bad outputs but split on borderline cases. Physician-validated uncertainty categories reveal that reducible uncertainty (missing context, ambiguous phrasing) more than doubles disagreement odds (OR = 2.55, p

CauKer: Classification Time Series Foundation Models Can Be Pretrained on Synthetic Data

arXiv:2508.02879v3 Announce Type: replace-cross Abstract: Time series foundation models (TSFMs) have recently gained significant attention due to their strong zero-shot capabilities and widespread real-world applications. Such models typically require a computationally costly pre-training on large-scale, carefully curated collections of real-world sequences. To allow for a sample-efficient pre-training of TSFMs, we propose \textsc{CauKer}, a novel algorithm designed to generate diverse, causally coherent synthetic time series with realistic trends, seasonality, and nonlinear interactions. \textsc{CauKer} combines Gaussian Process (GP) kernel composition with Structural Causal Models (SCM) to produce data for sample-efficient pre-training of state-of-the-art classification TSFMs having different architectures and following different pre-training approaches. Additionally, our experiments reveal that \textsc{CauKer}-generated datasets exhibit clear scaling laws for both dataset size (10K to 10M samples) and model capacity (1M to 783M parameters), unlike real-world datasets, which display irregular scaling behavior. The source code is publicly available at https://github.com/ShifengXIE/CauKer.

Stable Multi-Drone GNSS Tracking System for Marine Robots

arXiv:2511.18694v2 Announce Type: replace-cross Abstract: Stable and accurate tracking is essential for marine robotics, yet Global Navigation Satellite System (GNSS) signals vanish immediately below the sea surface. Traditional alternatives suffer from error accumulation, high computational demands, or infrastructure dependence. In this work, we present a multi-drone GNSS-based tracking system for surface and near-surface marine robots. Our approach combines efficient visual detection, lightweight multi-object tracking, GNSS-based triangulation, and a confidence-weighted Extended Kalman Filter (EKF) to provide stable GNSS estimation in real time. We further introduce a cross-drone tracking ID alignment algorithm that enforces global consistency across views, enabling robust multi-robot tracking with cooperative aerial coverage. We validate our system in diversified complex settings to show the accuracy and robustness of the proposed algorithm.

Meta-RL Induces Exploration in Language Agents

arXiv:2512.16848v2 Announce Type: replace-cross Abstract: Reinforcement learning (RL) has enabled the training of large language model (LLM) agents to interact with the environment and to solve multi-turn long-horizon tasks. However, the RL-trained agents often struggle in tasks that require active exploration and fail to efficiently adapt from trial-and-error experiences. In this paper, we present LaMer, a general Meta-RL framework that enables LLM agents to actively explore and learn from the environment feedback at test time. LaMer consists of two key components: (i) a cross-episode training framework to encourage exploration and long-term rewards optimization; and (ii) in-context policy adaptation via reflection, allowing the agent to adapt their policy from task feedback signal without gradient update. Experiments across diverse environments show that LaMer significantly improves performance over RL baselines, with 11%, 14%, and 19% performance gains on Sokoban, MineSweeper and Webshop, respectively. Moreover, LaMer also demonstrates better generalization to more challenging or previously unseen tasks compared to the RL-trained agents. Overall, our results demonstrate that Meta-RL provides a principled approach to induce exploration in language agents, enabling more robust adaptation to novel environments through learned exploration strategies.
❌