Citations

Abstract Background Lung adenocarcinoma in never-smokers is a major public health burden, especially among East Asian women. Polygenic risk scores (PRSs) are promising for risk stratification but are primarily developed in European-ancestry populations. We aimed to develop and validate single- and multi-ancestry PRSs for East Asian never-smokers to improve lung adenocarcinoma risk prediction. Methods PRSs were developed using genome-wide association study summary statistics from East Asian (8002 cases; 20 782 controls) and European (2058 cases; 5575 controls) populations. Single-ancestry models included a PRS with 25 single nucleotide variants (PRS-25), PRS clumping and thresholding (CT), and LDpred2; multi-ancestry models included LDpred2 plus European PRS-128, PRS-CSx, and clumping and thresholding, super learning, empirical Bayes (CT-SLEB). Performance was evaluated in independent East Asian data from the Female Lung Cancer Consortium and externally validated in the Nanjing Lung Cancer Cohort. We assessed predictive accuracy via area under the curve operating procedure (AUC), with 10-year and (ages 30-80 years) absolute risks estimates. Results The best multi-ancestry PRS, using East Asian and European data via CT-SLEB, outperformed the best East Asian-only (LDpred2; AUC = 0.629, 95% confidence interval [CI] = 0.618 to 0.641), achieving an AUC of 0.640 (95% CI = 0.629 to 0.653) and odds ratio of 1.71 (95% CI = 1.61 to 1.82) per standard deviation increase. Nanjing Lung Cancer Cohort validation confirmed robust performance (AUC = 0.649, 95% CI = 0.623 to 0.676). The top 20% PRS group had a 3.92-fold higher lung adenocarcinoma risk than the bottom 20%. Further, the top 5% PRS group reached a 6.69% lifetime absolute risk. Notably, this group reached the average population 10-year lung adenocarcinoma risk at age 50 years (0.42%) by age 41 years, 9 years earlier. Conclusions Multi-ancestry PRS approaches enhance lung adenocarcinoma risk stratification in East Asian never-smokers, with consistent external validation, suggesting future clinical utility.

ABSTRACT Objective The optimal treatment for neurosarcoidosis myelitis is uncertain. We characterize incident neurosarcoidosis myelitis and assess treatment response by MRI and clinical scales. Methods Incident probable or definite neurosarcoidosis myelitis in adults was retrospectively identified from 13 academic medical centers. Cases were analyzed by initial treatment. The primary outcome was T1 post‐contrast gadolinium enhancement resolution at 6 months post‐treatment. Secondary outcomes were changes in modified Rankin scale (mRS) and Expanded Disability Status Scale (EDSS) from nadir to final follow‐up. Results Two hundred two patients were identified (median diagnosis age: 47 years (IQR 39–55); male: female 1.3:1). Median nadir mRS and EDSS were 2 (IQR 2–3) and 4 (IQR 2.5–6). At initial treatment, 129 (63.9%) received prolonged corticosteroids ≥ 4 weeks (group A 1 ), 36 (17.8%) received corticosteroids < 4 weeks (B 1 ), 21 (10.4%) received corticosteroids plus sarcoidosis‐directed immunosuppressant (E), and 16 (7.9%) received corticosteroids plus non‐sarcoidosis‐directed agents (F). In 167 cases with sufficient imaging, there were no significant differences in contrast enhancement resolution at 6 months (A 1 27/106 (25.5%), B 1 9/28 (32.1%), E 5/19 (26.3%), F 5/14 (35.7%); Fisher's exact p = 0.76). There were no significant differences in changes in mRS or EDSS among treatment groups (Kruskal–Wallis p = 0.69 and 0.63, respectively) after median follow‐up of 46.5 months (IQR 18–91.3). Interpretation Different initial immunosuppression strategies did not correlate with MRI contrast enhancement resolution at 6 months or clinical scales (mRS, EDSS). However, conclusions are limited by retrospective design, imbalanced cohorts, and insensitivity of binary MRI outcomes and available clinical scales for treatment response in neurosarcoidosis.

Abstract Radiological protocol selection is a critical but time-consuming step in clinical workflow, requiring radiologists to match patient indications with the appropriate MRI or CT protocol. Manual selection can be prone to delays or potential errors, and automated approaches must contend with substantial data imbalance, site-specific variation, and evolving nomenclature. We investigated whether a large language model (LLM) can support reliable protocol selection at scale and whether retrieval-augmented generation (RAG) offers operational advantages over direct fine-tuning. Using 498,228 patient reports collected across three Mayo Clinic sites (Arizona, Florida, and Rochester) spanning six radiological divisions, we trained site-specific Llama 3.2 3B models for use with and without retrieval augmentation. Division-scoped Facebook AI Similarity Search (FAISS) indexes constructed from procedure and diagnosis text were used to supply contextual evidence in the RAG framework. Both fine-tuned and RAG-augmented models achieved strong performance across sites, with F1 scores of 0.88–0.90. RAG matched or modestly trailed direct fine-tuning overall but delivered consistent gains in specific divisions (e.g., musculoskeletal imaging). Importantly, the RAG model introduced abstention behavior (Not Enough Information), which concentrated in linguistically diverse divisions and provided an interpretable signal of uncertainty. These findings suggest that RAG-based models are viable for division-scoped protocol selection and offer practical advantages. Retrieval indexes can be refreshed far more easily and with fewer resources than retraining LLMs, enabling continual adaptation to evolving clinical workflows. Future prospective deployment will evaluate real-time accuracy, agreement with practitioners, and the role of abstention as a safety mechanism in clinical decision support.

Objective: Pulmonary air trapping is critical for diagnosing and prognostication of various lung diseases. Expiratory CT imaging serves as an accessible method to assess air trapping, which correlates with small airway disease outcomes. Air trapping manifests as mosaic attenuation on inspiratory chest CT that is difficult for visual estimation. The primary aim of this study was to develop an automated tool to quantify mosaic attenuation on inspiratory CT and air trapping on paired expiratory CT. Secondary aims included comparing CT-derived parameters with PFT measurements and dyspnea scores. Methods: This retrospective analysis of noncontrast chest CTs from 2 academic hospitals was conducted between January 1, 2018, and December 31, 2019. Patients with paired inspiratory and expiratory CT chest scans and PFTs performed on the same day were included. A chest radiologist manually annotated lung parenchyma in a reference cohort. Several histogram-based metrics were computed from lung parenchymal CT values, with the maximum peak position showing the strongest correlation with manually determined thresholds. This threshold, derived from the histogram peak, was applied in the adaptive thresholding process to quantify mosaic attenuation and air trapping. Results: We analyzed 267 patients (65.5% female, median age 68). Most exhibited normal physiological patterns (44.0%). Patients with elevated residual volume (RV) by PFTs (28.1%) had significantly higher inspiratory CT mosaic attenuation (1629.6 vs. 1311.5 mL, P <0.01) and expiratory CT air trapping volumes (1413.7 vs. 886.2 mL, P <0.01). Correlation analyses demonstrated strong relationships between CT-derived mosaic attenuation and air trapping measures and RV. The correlation with PFT parameters was even stronger in subgroup analyses in patients with obstructive PFT patterns. These models had good predictive ability for an abnormal RV (AUC of 0.92, sensitivity of 72.4%, and specificity of 92.0%) and clinical utility based on good correlation with the mMRC dyspnea score ( r =0.71; 95% CI: 0.65-0.77). Conclusions: This automated adaptive thresholding on inspiratory and expiratory chest CT scans showed a high correlation of lung volume and air trapping parameters with PFTs, revealing that measures of lung function have a complex interplay with air trapping.

Importance Advances in Alzheimer disease (AD) have shifted research focus to earlier disease stages, necessitating more scalable approaches to identify cognitively unimpaired individuals with amyloid β (Aβ) pathology. Objective To assess the utility of plasma phosphorylated tau 217 (p-tau217) for classifying Aβ status in cognitively unimpaired individuals, both as a stand-alone test and in a 2-step approach where positive plasma results were confirmed using a second modality (Aβ positron emission tomography [PET] or cerebrospinal fluid [CSF]). Design, Setting, and Participants This cross-sectional cohort study used data collected between June 2009 and March 2024. We included 2916 cognitively unimpaired participants from 12 international independent observational cohorts in the US, Europe, Australia, and Canada with available plasma p-tau217 levels and CSF or PET Aβ biomarkers. Performance comparisons between mass spectrometry and immunoassay-based p-tau217 measurements were also performed (n = 964). Exposures Plasma p-tau217 levels measured by immunoassay. Main Outcome and Measures Aβ status, determined by CSF or Aβ PET biomarkers. Results Participants had a mean (SD) age of 66.9 (9.9) years; 971 (33.3%) were Aβ positive by either CSF or PET, 1667 (57.2%) were women, and 1108 (38.1%) carried at least 1 APOE ε4 allele. As a stand-alone test, plasma p-tau217 achieved a positive predictive value (PPV) of 79% (95% CI, 74-84) and an overall accuracy of 81% (95% CI, 80-82). In a 2-step workflow, the PPV and accuracy significantly increased to 91% (95% CI, 86-95). While this approach required screening of 677 individuals with plasma p-tau217 to identify 100 Aβ-positive individuals, compared to 536 participants when using PET alone, it reduced the need for PET testing to 124. Immunoassays demonstrated comparable PPVs to mass spectrometry (80% [95% CI, 74-86] vs 85% [95% CI, 81-90]; P = .12) but significantly lower overall accuracy (82% [95% CI, 79-84]% vs 88 [95% CI, 86-90]; P &amp;lt; .001) and true Aβ-positive detection rate (49% [95% CI, 43-55] vs 69% [95% CI, 64-75]; P &amp;lt; .001). Conclusions and Relevance The findings highlight the potential of plasma p-tau217 as a stand-alone test—or when used in a sequential 2-step approach alongside PET or CSF testing—as a cost-effective, scalable, and minimally burdensome strategy for identifying preclinical AD. Tailored screening workflows that incorporate p-tau217 can improve efficiency in participant selection for preclinical AD trials and, in the future, help guide access to disease-modifying treatments.

Objective Closed-source large language models (LLMs) like generative pre-trained transformer 4o (GPT-4o) have shown promise for clinical information extraction but are potentially limited by cost, data security concerns and inflexibility. Open-source models are an attractive alternative with various adaptation strategies with no consensus on best practices. This study aims to rigorously identify optimal adaptation strategies for open-source models and evaluate their performance relative to closed-source alternatives. Methods and Analysis We studied three LLM adaptation strategies: chain-of-thought prompting, few-shot prompting and fine-tuning. Our target for information extraction was the Mayo Endoscopic Subscore (MES). We applied those strategies in all combinations to six open-source models (8–70 billion parameters) using an annotated set of colonoscopy procedure reports from the University of California, San Francisco (N=608) and San Francisco General Hospital (N=217). We analysed the relationship of these strategies to several performance metrics with a mixed-effects model, accounting for the variability between centres and LLMs. GPT-4o served as a closed-source oracle and provided in-depth commentary on the cost-effectiveness of these options. Results Quantised low-rank adaptation (QLoRA) statistically improves ( p §amp;lt; 0.001 ) ) the performance of open-source LLMs by 9.1–15.7 percentage points across accuracy, precision recall and annotation eligibility accuracy. However, GPT-4o with prompt engineering outperforms the best open-source model by 4.9%–11.2%. A simple cost-effectiveness analysis suggests that GPT-4o is more affordable compared with open-source alternatives. Conclusion GPT-4o is currently the most efficient LLM for MES extraction. If unavailable, QLoRA-optimised open-source models are a competitive alternative. However, results also suggest that current instruction-following LLMs including GPT-4o do not fully follow user-provided instructions, leaving room for improvement. More work is needed to achieve consistent, near-perfect performance in clinical information extraction by LLMs.

Abstract Electronic portal imaging devices (EPIDs) have been in widespread clinical use for almost two decades due to favorable measurement characteristics such as fast image acquisition, high sensitivity and resolution, digital data format, long‐term stability, linear dose response, and large detection area. On‐board EPIDs are widely used as a quality assurance (QA) tool to monitor nearly every aspect of linear accelerator (linac) performance. In both the commercial QA equipment market and in clinical practice, there is a trend toward replacing conventional QA approaches with fully automated, EPID‐based QA tools. Despite widespread use, EPIDs are complex devices, and there has yet to be consensus or formal recommendations published by the AAPM for their use in routine linac QA. Task Group 330 (TG‐330) was, therefore, formed to provide guidelines and recommendations for physicists on the safe and appropriate use of EPIDs for linac QA. In particular, the report: (a) provides a comprehensive review of the characteristics and limitations of EPIDs as time‐resolved measurement devices and dosimeters; (b) summarizes the application of EPIDs for linac QA; (c) provides recommendations on efficient and effective implementations of EPID‐based QA techniques; (d) describes risks associated with the use of EPIDs for linac QA and provides examples of how risk analysis can be used to ensure the safe use of EPIDs for linac QA. Many of the guidelines in this report are drawn from the literature that is included in the references and from the collective experience of the task group members. This report does not provide recommendations on the implementation of EPID‐based patient‐specific IMRT or VMAT QA, which is the topic of AAPM's report from AAPM TG‐307.