Citations

Retrieval-augmented generation (RAG) is rapidly emerging as a transformative paradigm for large language models (LLMs), especially in high-stakes domains like oncology that demand precision, factual grounding, and up-to-date knowledge. By pairing LLMs with external knowledge repositories, RAG systems explicitly ground model outputs in relevant retrieved documents, helping to reduce hallucinations and ensure responses reflect current evidence. In oncology, where clinical knowledge evolves continually with new research and drug approvals, RAG offers a way to integrate the latest data (e.g., trial results, guidelines, genomic databases) into decision-making. This review synthesizes the technical foundations of RAG, including its architecture and key components, and examines current applications in oncology such as clinical decision support, patient education, radiology reporting, pathology analysis, and genomics-driven precision medicine. We highlight recent studies that demonstrate RAG’s potential—for instance, improving treatment recommendations by incorporating genetic profiles and literature, and enhancing diagnostic accuracy by integrating guidelines. We also discuss emerging developments like multimodal RAG (combining text with imaging or other data), ensemble model approaches, and new explainability tools that trace model outputs to sources. Finally, we critically analyze the limitations and challenges of deploying RAG in healthcare, including computational costs, retrieval errors, noise or conflicts in retrieved information, and ethical and regulatory considerations. While RAG-based systems show promise in augmenting oncologists’ expertise with timely knowledge, careful implementation, high-quality curation of knowledge bases, and human oversight will be crucial for safe and effective adoption in clinical practice.

Background: Given the faster onset and shorter duration of action of ultra-rapid insulin lispro-aabc (URLi, Lyumjev™ insulin), compared with insulin lispro (a rapid-acting insulin [RAI]), the present study investigated the safety and glycemic outcomes of MiniMed™ 780G advanced hybrid closed-loop (AHCL) use with URLi in youths and adults with T1D. Methods: This 19-site, single-arm, clinical trial enrolled children and adolescents (aged 7–17 years) and adults (aged 18–80 years) with T1D who underwent a run-in period (∼2 weeks) using open loop or automated basal only with an RAI (aspart or lispro), followed by a 3-month study period using URLi with AHCL (AHCL + URLi). End-of-study glycemic metrics and A1c were compared to predefined thresholds and to data from baseline/run-in. Use of recommended optimal settings (ROS, 100 mg/dL glucose target with an active insulin time [AIT] of 2 h) was also assessed. Results: In an intention-to-treat population ( N = 101 pediatric and N = 110 adult participants), percentage of time in range (%TIR) during the last ∼6 weeks of the study period was significantly greater than the predefined thresholds for both the pediatric and adult participants. A1c was unchanged in the pediatric cohort but significantly reduced in the adult cohort (from 7.4 ± 0.9% to 6.9 ± 0.6% [ P < 0.001]). Compared with run-in, study period %TIR and percentage of time in tight target range (%TITR) were significantly increased (pediatric mean %TIRΔ [17.4%] and %TITRΔ [16.0%], respectively, and adult mean %TIRΔ [10.6%] and %TITRΔ [11.6%], respectively; P < 0.001 for all). Additionally, % time above range for both groups was significantly reduced. Use of ROS allowed more participants to achieve recommended glycemic targets. There were no severe hypoglycemia or diabetic ketoacidosis events. Most caregivers agreed that mealtime versus premeal URLi delivery simplified pediatric meal management. Conclusions: Similar to previous studies investigating automated URLi delivery, URLi use in the MiniMed™ 780G system was safe, and glycemic outcomes were favorable for pediatric and adult participants.

Background Failure of conventional imaging to detect pancreatic ductal adenocarcinoma (PDA) at its visually occult pre-diagnostic stage is a primary barrier to improving its otherwise poor rate of survival. Objective To develop and validate the Radiomics-based Early Detection MODel (REDMOD), an AI framework to identify subvisual radiomic signatures of pre-diagnostic PDA on standard-of-care CT. Designs REDMOD was trained on a multi-institutional cohort (n=969; 156 pre-diagnostic, 813 control) and tested on an independent set (n=493; 63 pre-diagnostic, 430 control), simulating a low prevalence (~1:6) early detection paradigm. The fully automated framework couples AI-driven segmentation with a heterogeneous ensemble architecture trained on a 40-feature radiomic signature derived from Synthetic Minority Over-sampling Technique (SMOTE)-balanced data. A tunable Youden Index-optimised classification threshold enables performance calibration without retraining. Validation included direct comparison with radiologists, longitudinal test–retest analysis and external specificity validation across two independent cohorts (n=539 and n=80). Results On an independent test set (n=493), REDMOD identified occult PDA (AUC 0.82; 73.0% sensitivity) at a median 475-day lead time. This represented nearly twofold higher sensitivity than radiologists (38.9%; p<0.001), which grew to nearly threefold (68.0% vs 23.0%) at >24 months lead time. REDMOD showed strong longitudinal stability (90–92% concordance) and generalisable specificity across multi-institutional (81.3%; n=539) and public (87.5%; n=80) datasets. Mechanistic analyses confirmed predictive power derived principally from multi-scale wavelet-filtered textural features (90% of selected signature), which outperformed unfiltered features (AUC 0.82 vs 0.74; p=0.007) in capturing subvisual architectural disruptions. Conclusions REDMOD is an automated, mechanistically grounded, longitudinally stable, externally validated AI that surpasses radiologists for PDA detection at its visually occult pre-diagnostic stage. These attributes position it for prospective validation in high-risk cohorts, a necessary step towards shifting the paradigm from late-stage symptomatic diagnosis to proactive pre-clinical interception.

Abstract Background Presacral tumours are rare and heterogeneous lesions arising within a complex anatomical space, resulting in variability in diagnosis and management. Increasing incidental detection and evolving surgical and oncological strategies have highlighted the need for contemporary guidance. The aim of this study was to establish an international expert consensus on the optimal diagnosis and management of presacral tumours. Methods An international panel of colorectal surgeons from high-volume units was convened. Evidence- and practice-based statements were developed after domain-based literature reviews and refined through five online and two in-person meetings. Consensus was achieved using a modified Delphi process with anonymous voting. Results In total, 16 colorectal surgeons participated in the study. Ten domains were identified resulting in the development of 22 statements based on expert opinion and retrospective studies. After two rounds, unanimous (100%) agreement was achieved for all statements. High-resolution MRI with structured synoptic reporting is endorsed as central to risk stratification. Selective image-guided transperineal or parasacral biopsy is recommended for high-risk lesions where histology would alter management. Complete surgical resection is defined as the operative goal, with approach tailored to tumour biology and anatomy; minimally invasive surgery is considered appropriate in selected cases. Non-operative surveillance is supported for carefully selected asymptomatic, low-risk cystic lesions. Postoperative surveillance is recommended in a risk-adapted manner according to histopathology and margin status. Conclusion This international consensus provides a pragmatic, risk-stratified framework for the multidisciplinary diagnosis and management of presacral tumours. These recommendations aim to reduce practice variation, support decision-making in specialized centres, and inform future prospective collaborative research.

Importance Seizure-induced brain injury is central to the treatment urgency of new-onset refractory status epilepticus (NORSE). Identifying biomarkers that reflect ongoing neuronal damage could inform therapeutic timing and improve outcomes. Objective To quantify acute brain injury in patients with cryptogenic NORSE (cNORSE), etiology-defined status epilepticus (eSE), and chronic epilepsy. Design, Setting, and Participants This was an international cross-sectional study conducted between 2013 and 2025. Patients were enrolled at 36 hospitals in the US, 2 in Canada, and 1 in Italy, France, and Belgium. Patients with cNORSE and eSE for which biological samples were obtained during ongoing seizure activity were enrolled in the study. Comparison groups without status epilepticus comprised individuals with chronic epilepsy and healthy participants. None were excluded. Exposures Neurofilament light chain (NfL) and S100-beta (S100B) protein concentrations in serum and cerebrospinal fluid (CSF). Main Outcomes and Measures Degree of neuronal and glial damage, indexed by NfL and S100B levels, and their association with short-term functional outcomes. Results A total of 78 patients with cNORSE (mean [95% CI] age, 37 [30-41] years; 44 female [56%]) and 2 independent cohorts of 211 patients (mean [95% CI] age, 69 [66-71] years; 128 female [61%]) and 73 patients (mean [95% CI] age, 56 [45-65] years; 39 male [53%]) with eSE were included. NfL concentrations were markedly elevated in cNORSE—approximately 10-fold higher in CSF and 4-fold higher in serum—compared with the eSE cohorts (CSF: median [IQR], 6408 [1503-22 963] pg/mL compared with 694 [219-2389] pg/mL; serum: median [IQR], 231 [99-855] pg/mL compared with 55 [20-135] pg/mL; P &amp;lt;.001). Serum NfL levels were nearly 20-fold higher in cNORSE than in the cohort with epilepsy and in healthy controls (median [IQR], 11 [7-19 ] and 7 [5-14 ] pg/mL, respectively). Serum and CSF NfL levels were strongly correlated (Spearman ρ = 0.75; P &amp;lt; .001) and rose sharply between week 1 (median [IQR], 101 [51-137] pg/mL), week 2 (median [IQR], 197 [117-324] pg/mL), and week 3 (median [IQR], 598 [163-1000] pg/mL) after onset ( P &amp;lt; .001). In contrast, S100B concentrations did not differ between groups and showed no consistent temporal pattern. NfL discriminated cNORSE from eSE (area under the receiver operating characteristic curve [AUROC], 0.79; 95% CI, 0.68-0.90) and from cohorts without status epilepticus (AUROC, 0.99; 95% CI, 0.78-1.00). Higher serum NfL was independently associated with poor functional outcome at discharge (Glasgow Outcome Scale extended score, 1-4; odds ratio, 1.01; 95% CI, 1.00-1.03; P = .03). Conclusions and Relevance Results of this cross-sectional study suggest that acute neuroaxonal injury, as reflected by elevated NfL levels, was substantially greater in cNORSE than in the cohorts with eSE and in controls without status epilepticus. The rapid early rise in NfL highlights a narrow therapeutic window, emphasizing the need for prompt, effective, and potentially neuroprotective interventions in cNORSE.

INTRODUCTION: Nosocomial infections (NIs) in cirrhosis are associated with high mortality but could be preventable. Logistic regression (LR) models have failed to identify high-risk patients. We aimed to develop machine learning (ML) models to predict NI. METHODS: The CLEARED consortium consists of prospectively enrolled cirrhosis inpatients from >120 centers. Using day-of-admission clinical data, 3 ML approaches (random forest [RF], extreme gradient boosting, and neural networks [NNs]) were used to predict NI. Data were split 80:20 for training and testing stratified by the outcome. Models were compared using area under the receiver operating characteristic curve (AUC). RESULTS: In total, 8,263 patients (55.90 ± 13.34 years; 64.1% men) from 127 centers in 37 countries were included. NI developed in 869 (10.5%), a median of 6 (4–11) days of postadmission. Major NIs were respiratory (29.6%) and urinary tract infection (15.7%), and spontaneous bacterial peritonitis (13.5%). NIs occurred more frequently in patients from low/low-middle income countries and those with severe liver disease, alcohol etiology, and admission infections. NIs were associated with inpatient mortality (31.9% vs 8.1%, P < 0.001) and liver transplantation (4.8% vs 1.9%, P < 0.001). Although the RF model (AUC 0.69) showed good calibration (Brier score 0.09), outperforming extreme gradient boosting, neural network, and LR models (AUC 0.66 for all; LR comparison P = 0.043), no model achieved AUC ≥0.80 for clinical utility. At 10% predicted probability threshold, the RF model demonstrated only 75.4% sensitivity, 52.9% specificity, and 15.9% positive predictive value (PPV). DISCUSSION: NIs cannot be accurately predicted from day-of-admission data using ML models, even in a large, prospective, global cirrhosis cohort. Every hospitalized patient with cirrhosis should receive protocolized infection control measures.

ABSTRACT In 2018, the International Niemann‐Pick Disease Alliance (INPDA) and the International Niemann‐Pick Disease Registry (INPDR) developed and published comprehensive clinical management guidelines to support inclusive and standardized care pathways in Niemann‐Pick disease type C (NPC)—an ultra‐rare, autosomal recessive, neurovisceral lysosomal disorder. Since then, advances in diagnostics, care, and the approval of two novel disease‐modifying agents have underscored the need to revise these guidelines to ensure safe, consistent, and high‐quality care for those affected by NPC. In response, the INPDA and INPDR convened a multidisciplinary Guidelines Development Group (GDG) comprising individuals with NPC expertise from 14 countries across five continents, representing a broad range of specialties, as well as patients and families involved in NPC care. Informed by a comprehensive literature review and two meetings, the GDG systematically reviewed, revised, and updated the 2018 guideline statements, re‐evaluating the level of evidence, strength of recommendations, and expert agreement for each. The resulting 2025 consensus clinical management guidelines constitute a timely, up‐to‐date, and internationally applicable resource for the diagnosis, treatment, and holistic management of individuals with NPC. These guidelines serve as a critical resource for specialist centers, hospital‐based medical teams, staff involved in NPC patient care, family physicians and other primary caregivers, and, importantly, patients and their families.

Abstract Parkinson’s disease (PD) is commonly associated with dysfunctional mitochondrial homeostasis. PINK1, a S/T kinase mutated in early-onset PD, generates phosphoserine 65 ubiquitin (pS65Ub) on damaged mitochondria facilitating their removal. Here, we show that pS65Ub translocates into the nucleus after generation at damaged mitochondria and is directly attached to substrates by resident E3 ligases. Histone H2A is a major substrate and is modified at lysine 119 (H2AK119) by the polycomb silencer, E3 ligase RING1B. At nucleosomes, pS65Ub simultaneously suppresses RING1B and potentiates H2A deubiquitinases USP16 and USP21. Epigenetic profiling and RNA sequencing reveal that pS65Ub is enriched at the promoters of poorly expressed yet dynamically regulated genes and is associated with H2AK119ub depletion. Functionally, we show that pS65Ub enrichment drives polycomb target gene expression, which accelerates the maturation of dopaminergic neurons. Importantly, post-mortem PD brains exhibit elevated nuclear pS65Ub, potentially linking nuclear pS65Ub accumulation with disease pathogenesis. Together, these data indicate that pS65Ub generated at damaged mitochondria regulates fundamental cellular processes at distant sites.