Citations

Abstract Cancer outcomes vary widely between individual patients, each accumulating an irregular record of treatments, diagnoses, measurements, and complications. Current prognostic models reduce this complexity into a single snapshot, focus on narrow clinical settings, and rarely generalize across hospitals. Here we introduce Chronicle , an explainable transformer that learns from entire patient trajectories to predict diverse clinical outcomes throughout the disease course while capturing both short- and long-term temporal dependencies. Trained on 53.7 million longitudinal data points from 51,711 patients spanning 67 cancer types, Chronicle operates natively on irregular data without imputation and jointly predicts eight endpoints within a flexible framework adaptable to additional outcomes. Chronicle outperformed cross-sectional models for overall survival prediction (C-index 0.84 vs 0.76-0.79), stratified patients more accurately than established prognostic systems, including TNM stage, and predicted seven adverse event and transfusion endpoints (AUC 0.80-0.92). Applied without retraining to 69,341 patients in Germany, Switzerland, and the United States, Chronicle generalized across healthcare systems and improved further with local fine-tuning. Integrated explainability traced each risk update to patient-specific clinical factors, revealing distinct temporal persistence of prognostic information, with relevance half-lives ranging from weeks for therapies to nearly one year for baseline characteristics. These findings demonstrate that learning from hospital-wide patient trajectories enables interpretable and continuously updated predictions, providing a scalable framework to support individualized treatment decisions.

Background Randomized trials have shown equivocal outcomes between a direct aspiration as first-pass technique (ADAPT) and stent retriever (SR) stroke thrombectomy, yet whether specific patient subgroups benefit differentially from one technique over the other remains unknown. Methods In this retrospective multicenter cohort study, patients with large vessel occlusion were included, and those who underwent ADAPT versus SR first-pass thrombectomy (either as standalone or combined with aspiration) were compared. The primary outcome was a composite of first-pass effect (FPE, modified Thrombolysis In Cerebral Infarction (mTICI) 2c or better with one attempt) with no symptomatic intracranial hemorrhage (sICH). A causal forest model was used to estimate the conditional average treatment effect (CATE) for each patient, with larger CATEs indicating that ADAPT is expected to be more advantageous. Patients were stratified by CATEs, and outcomes were compared between treatment groups. Results Among 6700 included patients, ADAPT (n=3695, 55%) was associated with a higher rate of FPE with no sICH (39.9% vs 26.9%; adjusted OR (aOR) 2.07, 95% CI 1.77 to 2.41, P<0.001). The causal forest identified significant treatment effect heterogeneity (interaction P<0.001), where ADAPT was more advantageous in more recent years and among patients with M1 occlusions presenting in earlier time windows. Low-CATE patients showed no statistically significant difference between techniques (aOR 0.91, 95% CI 0.57 to 1.45, P=0.70), while medium- and high-CATE patients demonstrated progressively larger ADAPT advantages (aOR 2.06, 95% CI 1.67 to 2.55, and aOR 2.99, 95% CI 2.24 to 3.98, respectively, both P<0.001). Conclusions ADAPT may be superior to SR for a majority of patients, particularly in recent years, and among patients with M1 occlusions presenting in earlier time windows.

Background The management of osteochondral lesions of the talus (OLT) is challenging, with 2 main options being conservative and surgical management. Orthobiologic therapies come to bridge the gap between these 2 options; however, inconsistencies in indication, composition, and heterogeneous evidence limit clinical decision-making. To address these challenges, 3 International Congress Meetings on Cartilage Repair of the Ankle were convened in Pittsburgh, USA (2017), Dublin, Ireland (2019), and New York, USA (2025) to establish agreement on the role of conservative management and injectable orthobiologic strategies in the treatment of OLT. Methods Seventy-five multidisciplinary experts from over 25 countries and 1 territory participated in a structured Delphi-based consensus process. Blinded surveys, structured questionnaires, and in-person discussions were used to develop and refine statements, supported by literature review and graded evidence (A1–E). Consensus strength was defined as consensus (51%-74%), strong consensus (75%-99%), or unanimous (100%) Results In total, 29 consensus statements were achieved across the 3 meetings. Strong agreement was reached regarding indications and expected outcomes of conservative management, including the role of immobilization and appropriate timing for surgical intervention. Orthobiologic injections, including platelet-rich plasma, concentrated bone marrow aspirate, and hyaluronic acid, were considered appropriate in select patients with persistent symptoms, although no superiority was demonstrated among formulations, preparation methods, or injection strategies. When used as adjuncts to surgical procedures, orthobiologics may improve radiological outcomes, with limited evidence for improvement in clinical outcomes. Technical considerations, including preferred harvest sites and a lack of standardized preparation protocols, were identified, along with key principles for patient counseling Conclusion These international consensus statements establish a standardized framework for the conservative management and use of injectable orthobiologic therapies in OLT. By clarifying indications, technical considerations, and current limitations, this work aims to improve consistency in clinical practice, guide treatment decision-making, and inform future research.

Radiation oncology is poised for a decade of unusually rapid technological and biological change. Six innovations are converging on the clinic: FLASH radiotherapy, adaptive radiotherapy with advanced imaging, artificial intelligence and machine learning in treatment planning, theranostics with targeted radionuclide therapy, spatially fractionated radiotherapy using GRID and LATTICE techniques, and continuing enhancements in particle therapy with protons and heavy ions. Each is biologically compelling, but they will not reach patients at the same speed or in equal numbers. This review proposes an evidence-and-capital framework for evaluating these innovations along four axes – physical principle, biological rationale, current clinical evidence, and dominant adoption barrier – and ranks them by anticipated population-level impact over the next five years. Capital intensity emerges as the principal determinant of diffusion: innovations layered onto existing linear accelerators (artificial intelligence, spatially fractionated radiotherapy) or onto existing nuclear medicine infrastructure (theranostics) will scale globally, whereas innovations requiring new bunkers, magnetic-shielded vaults, or dedicated accelerators (adaptive radiotherapy, FLASH, particle therapy) will remain concentrated in well-resourced centres. A low anticipated diffusion ranking is not a judgement on intrinsic clinical value: several capital-intensive modalities deliver substantial benefit for selected patients even where population-level reach is limited, and the framework distinguishes throughout between intrinsic efficacy and expected accessibility. The resulting equity gap is not primarily a scientific problem but an infrastructural one. Recognising this asymmetry should inform departmental investment, training curricula, multinational cooperative-group trial design, and health-system policy on capital subsidy and consortium models. The framework is intended for radiation oncologists, medical physicists, and cancer scientists evaluating where to invest finite institutional resources.

ABSTRACT Introduction Cardiogenic shock (CS) carries high morbidity and mortality. However, evidence for mechanical circulatory support (MCS) in non‐ischemic CS remains limited, as prior trials have focused on infarct‐related shock. Methods We performed a retrospective cross‐sectional study using the Nationwide Inpatient Sample (2017–2019), identifying hospitalizations with a principal diagnosis of CS and excluding acute myocardial infarction. Patients were categorized into mutually exclusive groups: no MCS, intra‐aortic balloon pump (IABP) only, Impella only, or extracorporeal membrane oxygenation (ECMO) only. Primary outcomes were in‐hospital mortality, length of stay (LOS), and total hospital cost. Survey‐weighted multivariable models adjusted for demographics, hospital characteristics, and comorbidity burden using Elixhauser and Charlson indices. Results Among 104 045 hospitalizations, 6040 received MCS. IABP was most common, followed by Impella and ECMO. Among patients receiving a single MCS strategy, overall in‐hospital mortality was 29.5%., with lower mortality in the IABP group (19.9%) compared with Impella (37.4%) and ECMO (44.3%). After adjustment, IABP was associated with lower observed in‐hospital mortality versus no MCS (aOR 0.58, 95% CI 0.53–0.64; p < 0.01), whereas Impella and ECMO showed increased aOR. IABP and ECMO were associated with longer LOS, while Impella was not. All MCS strategies were associated with higher costs, with ECMO showing the greatest increase (cost ratio 3.30, 95% CI 3.12–3.50). Conclusion In non‐ischemic CS, MCS strategies demonstrate distinct profiles of mortality and resource utilization. IABP was associated with lower observed in‐hospital mortality, whereas all MCS were linked to higher costs, highlighting heterogeneity in outcomes and the need for further research.

ABSTRACT Beta‐lactam antibiotics (beta‐lactams) are first‐line treatments for most major infectious syndromes in acutely ill patients, with in vitro activity against common pathogens, demonstrated efficacy in trials, and a perceived low risk of adverse effects. The current dosing approach, informed by population‐level data, tends to be simplistically reduced to a “one size fits all” dosing method which only accounts for body weight (i.e., in pediatrics) and end organ function to estimate drug clearance (e.g., estimated glomerular filtration rate); this approach results in substantial variability in observed serum concentrations in acutely ill patients, which can compromise real‐world effectiveness and safety. Beta‐lactam dose individualization, therefore, defined as a dosing regimen for one patient informed by measured concentrations from that patient, has been recommended in international clinical guidelines. Comprehensive guidance on best practices for beta‐lactam dose individualization is lacking. To address this knowledge gap and develop guidance, a multidisciplinary panel of international experts was assembled from the disciplines of infectious diseases, critical care, pharmacometrics, and laboratory medicine, representing both adults and pediatrics. The panel systematically evaluated the literature, using the GRADE approach where feasible, to address broad thematic questions pertaining to whether beta‐lactam dose individualization should be pursued, for what indications beta‐lactam dose individualization is warranted, and the most critical considerations and best practices for implementation of beta‐lactam individualization. The resultant consensus recommendations will equip healthcare professionals caring for acutely ill patients treated with beta‐lactam therapy with the evidence and tools necessary to guide optimal use of beta‐lactam dose individualization.

Abstract Frontotemporal lobar degeneration (FTLD) and Alzheimer’s disease (AD) differ in their clinical features and genetic etiologies but share progressive cognitive decline. Emerging evidence implicates lipid dysregulation in neurodegeneration, but its extent across FTLD subtypes and how it compares to AD are unclear. Here, we performed integrated lipidomic and proteomic analyses of matched frontal (disease-vulnerable) and occipital (relatively spared) post-mortem cortices from individuals with genetic and sporadic FTLD-TDP, FTLD-tau (Pick’s disease, PiD), AD, and controls. FTLD and AD exhibited convergent lipid alterations, including reduced levels of cardiolipins and phosphatidylethanolamines, alongside increased gangliosides, diacylglycerols, cholesterol esters, acylcarnitines, and coenzyme Q, with generally greater changes in FTLD frontal cortex. FTLD displayed additional alterations, including reductions in bis(monoacylglycerol)phosphate, ceramides, phosphatidylserines, phosphatidylinositols, and sulfatides. These lipid changes were accompanied by proteomic alterations involving lysosomal proteins, phospholipases, phospholipid remodeling enzymes, and fatty acid oxidation pathways. Although lipidomic and proteomic signatures were broadly shared across FTLD subtypes, GRN associated FTLD-TDP and PiD showed the most extensive alterations. Triglycerides were selectively reduced in PiD in association with decreased DGAT1 expression, whereas cholesterol esters were elevated across all subtypes except C9orf72 associated FTLD-TDP. These findings identify shared disruptions in lipid homeostasis and lysosomal lipid metabolism across FTLD and AD, highlighting convergent metabolic pathways underlying neurodegeneration.

BACKGROUND: White matter hyperintensities (WMH) are widely used to assess cerebral small vessel disease but reflect late-stage injury. Diffusion magnetic resonance imaging (MRI) biomarkers have been proposed to capture earlier small vessel disease-related microstructural damage but their temporal progression relative to WMH and risk factors associated with progression remain unexplored. METHODS: We identified 2077 participants from the population-based cohort study of the Mayo Clinic Study of Aging in Olmsted County, Minnesota (aged 50–101 years) collected between 05/2005 and 09/2024 with longitudinal neuroimaging. Using multioutput nonlinear mixed-effects models in those with at least 2 fluid-attenuated inversion recovery-MRI and diffusion MRI scans, we characterized the temporal progression of WMH and 4 diffusion MRI biomarkers: fractional anisotropy of the genu of the corpus callosum, peak width of skeletonized mean diffusivity, free water, and Arteriolosclerosis-score, which were automatically estimated. Models incorporated participant-specific time shifts, correlations between biomarkers, and effects of risk factors (sex, education, APOE ε4 status, cardiometabolic conditions). RESULTS: The study population had a mean age of 78 years, 47% were women, 28% were APOE ε4 allele carriers, and 80% were cognitively unimpaired, with an average follow-up of 5.2 years (SD, 4.3 years) for fluid-attenuated inversion recovery-MRI and 4.3 years (SD, 3.9 years) for diffusion MRI. Arteriolosclerosis-score, fractional anisotropy of the genu of the corpus callosum, free water, and peak width of skeletonized mean diffusivity became abnormal in 50% of the study population 16, 12, 10, and 7 years before WMH become abnormal (half-width of CI <1 year), respectively. Global markers (Arteriolosclerosis-score, free water, peak width of skeletonized mean diffusivity, and WMH) were correlated, indicating shared substrates of widespread white matter injury. Fractional anisotropy of the genu of the corpus callosum, a vascular risk microstructural injury biomarker, was weakly coupled with WMH and had an earlier but more linear worsening across adulthood. Cardiometabolic conditions predicted earlier worsening of all biomarkers. Female participants showed earlier WMH, fractional anisotropy of the genu of the corpus callosum, and Arteriolosclerosis-score abnormalities, whereas male participants exhibited earlier peak width of skeletonized mean diffusivity and free water abnormalities. CONCLUSIONS: Diffusion MRI biomarkers were abnormal at least a decade before WMH become abnormal in the population, revealing a prolonged phase of early small vessel disease and highlighting their potential for small vessel disease prevention.

BACKGROUND Malignant peripheral nerve sheath tumors (MPNSTs) are malignant tumors that rarely demonstrate heterologous differentiation from angiosarcomatous components. These lesions can be hard to accurately diagnose. OBSERVATIONS A 25-year-old man with neurofibromatosis type 1 and a history of a left lumbar subcutaneous MPNST developed a lung metastasis of only the angiosarcoma component of the MPNST. Prior to H3K27me3 staining of the lung mass, there was no association of the angiosarcoma with the MPNST. Reinterpretation of the lumbar subcutaneous MPNST led to the identification of a previously unrecognized angiosarcomatous portion in that tumor as well. LESSONS This case highlights the importance of using H327me3 staining for accurate diagnosis of biopsies only capturing the heterologous components of MPNST. https://thejns.org/doi/10.3171/CASE251012

BACKGROUND Cystic adventitial disease (CAD) and intraneural ganglion cysts (IGCs) are rare mucinous lesions typically linked to an articular origin via macroscopic conduits. While secondary propagation between vascular and neural systems may occur through shared fascial sheaths, a direct microvascular route has not been previously described. OBSERVATIONS A 60-year-old man presented with progressive left tibial neuropathy. High-resolution 7T MRI and surgery revealed a large popliteal CAD lesion connecting to the knee joint via the middle genicular artery, alongside a secondary tibial IGC lacking an independent articular nerve connection. Findings demonstrated that high-pressure CAD mucin tracked along a small connecting vascular branch, creating a novel microvascular arterio-neural pathway to infiltrate the tibial nerve. Targeted articular branch disconnection and cyst decompression yielded significant clinical improvement, with no recurrence on 7-month follow-up 7T MRI. LESSONS This case expands the unifying articular theory, demonstrating that synovial fluid propagation can exploit a microvascular arterio-neural pathway rather than strictly macroscopic fascial planes. High-resolution 7T MRI is essential for visualizing these subtle connections. Successful management requires meticulous identification and disconnection of the primary joint connection to prevent recurrence. https://thejns.org/doi/10.3171/CASE26411