Citations

Abstract Background Accurate and efficient dose calculation is essential for online adaptive planning in proton therapy. Deep learning (DL) has shown promising dose prediction results for pencil beam scanning proton therapy (PBSPT) in recent years, but existing DL‐based dose prediction methods still suffer from limited generalizability and an inability to effectively handle outlier clinical cases. This may lead to inaccurate dose delivery to targets or excessive irradiation to organs at risk (OARs), thereby compromising the safety and efficacy of online adaptive proton therapy. Purpose To design a physics‐aware and generalizable AI‐based PBSPT dose prediction method that incorporates underlying physics to enhance generalizability, particularly in handling outlier clinical cases. Methods This study analyzed PBSPT plans of 103 prostate (93 for training and 10 for testing) and 78 lung cancer patients (68 for training and 10 for testing) from our institution, with each case comprising CT images and structure sets. Using the doses generated by our Monte Carlo‐based dose engine as the reference standard, we compared three methods: the region of interest (ROI)‐based method, the beam mask and sliding window method, and the proposed noisy probing dose method, which rapidly generates a low‐statistics dose via uniformly weighted spots on an expanded spot‐placement target volume without optimization. To evaluate the generalizability of these methods to rare treatment planning scenarios, 12 cases with uncommon beam angles or prescription doses were used to assess their performance, which was evaluated using dose‐volume histogram (DVH) indices, 3D Gamma passing rates (3%/2 mm/10%), and Dice coefficients for dose agreement, while prediction times were measured to gauge model efficiency. Results The proposed noisy probing dose method consistently outperformed the ROI‐based and beam mask baselines across all evaluated metrics, with more accurate dose agreement and superior generalizability. For DVH indices, the noisy probing dose method achieved the smallest deviation in clinical target volume (CTV) dose coverage: in prostate cancer, CTV D98 deviation was reduced by 45% (from 0.53 ± 0.22 Gy [RBE] for ROI‐based) and 29% (from 0.41 ± 0.28 Gy [RBE] for beam mask) to 0.29 ± 0.06 Gy [RBE]; in lung cancer, similar improvements were observed, with CTV D98 deviations reduced to 0.34 ± 0.12 Gy [RBE]. The 3D Gamma passing rates improved to 99.65% ± 1.15% for prostate targets and 97.04% ± 1.17% for lung targets. The dice coefficients of the 90% iso‐dose lines were also the highest with the noisy probing dose method (prostate: 0.983 ± 0.005; lung: 0.967 ± 0.01). For the 12 outlier cases, the noisy probing dose method maintained superior generalizability, yielding higher 3D Gamma passing rates (prostate targets: 96.79% ± 0.83%, OARs: 94.29% ± 1.01%; lung targets: 93.38% ± 1.34%, OARs: 93.95% ± 1.32%), demonstrating robust generalizability to rare clinical scenarios. The dose predictions for all testing cases were completed within 0.3 seconds. Conclusions A novel noisy probing dose method was proposed for PBSPT dose prediction in prostate and lung cancer patients. By embedding more proton‐specific physics, this method demonstrated an improvement in the generalizability of dose prediction.

ABSTRACT Objective Management of vestibular schwannoma (VS) remains controversial. Very small VSs represent a pivotal group: on one hand, they pose the least risk to the patient from the perspective of being small; yet, not recommending treatment may result in missing a critical window for achieving the best outcome. This debate fundamentally hinges on whether treatment improves upon the natural history of the disease. There is an unproven assumption that very small tumors behave similarly to the broader observed population. To address this central question, we conducted a multicenter retrospective review to compare the natural history of tumor growth and hearing for patients with very small VS (1–3 mm) with a referent cohort with larger tumors. Methods All patients diagnosed with sporadic VS 1–3 mm were reviewed at 3 large skull base referral centers and compared with a referent cohort of patients with tumors > 3 mm. Tumor growth‐free survival rates, treatment‐free survival rates, and rates of maintenance of class A, B, or C hearing were estimated using the Kaplan–Meier method. Results In total, 123 patients with very small VS and a referent cohort of 326 patients were studied. Seventeen patients with very small VS experienced tumor growth at a median of 3.2 years after diagnosis (IQR 2.6–5.1), demonstrating a median growth rate of 1.0 mm per year (IQR 0.9–1.8). In comparison, 167 of the patients in the referent cohort experienced tumor growth at a median of 1.7 years after diagnosis (IQR 0.8–3.1). Tumor growth‐free survival rates (95% CI, number still at risk) at 5 years after diagnosis for the 2 cohorts were 84% (75–93, 39) and 45% (40–52, 100), respectively ( p < 0.001). Six patients with very small VS underwent treatment after tumor growth at a median of 3.2 years after diagnosis (IQR 3.2–3.8), compared with 138 patients in the referent cohort at a median of 2.2 years after diagnosis (IQR 1.4–3.6). Treatment‐free survival rates (95% CI, number still at risk) at 5 years after diagnosis for the 2 cohorts were 92% (86–98, 55) and 54% (49–61, 125), respectively ( p < 0.001). Among the 105 patients with very small VS and class A, B, or C hearing at diagnosis, 3 progressed to class D. Comparatively, among the 225 patients in the referent cohort with class A, B, or C hearing at diagnosis, 70 progressed to class D at a median of 2.6 years after diagnosis (IQR 1.0–4.8). Rates of maintenance of class A, B, or C hearing (95% CI, number still at risk) at 5 years after diagnosis were 96% (92–100, 28) for patients with very small VS and 65% (57–73, 70) for the referent cohort ( p < 0.001). Conclusions The natural history of tumor growth and hearing for very small VS diverges favorably from the cumulative natural history data and provides a more appropriate benchmark to compare clinical outcomes of upfront surgery or radiosurgery for this population. At 5 years after diagnosis, 96% of very small VS maintained class A, B, or C hearing and only 8% received treatment for growth within this multicenter cohort. Cumulatively, these data substantiate initial wait‐and‐scan management via size threshold surveillance in this population. Level of Evidence 3.

Abstract Heterogeneous treatment effects (HTE) are key to precision medicine, but most real-world studies lack the scale and diversity needed to detect them. While multi-site analyses offer a potential solution, data-sharing constraints often prevent access to patient-level information across institutions. We introduce POLARIS, a federated framework for time-to-event target trial emulation. POLARIS converts each site’s weighted Cox risk function into a compact tensor shared once with the coordinating center, enabling lossless reproduction of pooled estimates without sharing patient-level data. We applied POLARIS across five U.S. health systems to study risk of gastrointestinal outcomes after GLP-1 receptor agonist (GLP-1RAs) initiation versus sodium-glucose cotransporter 2 inhibitors (SGLT2is) and dipeptidyl peptidase 4 inhibitors (DPP4is). Results showed that GLP-1RAs were consistently associated with higher risks of nausea and vomiting, particularly among men, individuals with higher baseline HbA1c (≥ 8.5%), and lipid therapy. POLARIS provides a scalable solution for distributed target trial emulation and fine-grained assessment of HTE across diverse health systems.

Expert consensus recommendations from 21 academic centers were established for appropriate radiography utilization at 12 potentially low-yield anatomic sites through a multidisciplinary process endorsed by six major American medical societies.

As an emerging tire wear-derived environmental contaminant, 6PPD-quinone (6PPDQ) has raised significant concerns regarding its neurotoxic potential, particularly for children exposed to recycled tire crumb rubber in playgrounds. However, the molecular mechanisms by which 6PPDQ influences neurological disorders such as epilepsy remain poorly understood. In this study, we employed an integrative framework combining network toxicology, bulk analysis of human epileptic brain tissues, Mendelian randomization, and molecular dynamics simulations to elucidate these mechanisms. Our findings, validated through CETSA-WB and SPR, identify 6PPDQ as a direct ligand that binds to and stabilizes neuronal TP53. Through a synergistic double-hit mechanism, 6PPDQ directly engages the TP53 pathway while simultaneously triggering microglial interleukin-6 secretion. These converging pathways lead to the suppression of the master antioxidant regulator Nrf2, resulting in glutathione depletion, excessive reactive oxygen species accumulation, and exacerbated neuronal damage under excitotoxic stress. Experimental validation using glutamate-induced HT22 cell models and microglia–neuron crosstalk systems confirmed that targeting the TP53/Nrf2 axis or scavenging ROS significantly attenuates 6PPDQ-induced neurotoxicity. Our findings highlight critical risks to pediatric neurological health posed by tire-derived contaminants and identify the TP53/Nrf2 axis as a promising therapeutic target. Furthermore, this work provides a robust scientific basis for refining risk assessment frameworks and developing regulatory strategies to mitigate environmental exposure to 6PPDQ.

Abstract Background Acute kidney injury (AKI) frequently causes remote organ injury including hepatic steatosis, yet whether lipid accumulation reflects increased synthesis or impaired clearance has not been resolved. Methods We used a murine ischemia-reperfusion AKI model. Unbiased liver proteomics was performed at 24 hours after reperfusion, and dysregulated pathways were identified by Gene Set Enrichment Analysis. Results were validated by Western blotting, qPCR, and immunohistochemistry. These findings were complemented by retrospective analysis of two ICU databases (MIMIC-IV and eICU-CRD). Results AKI significantly increased serum ALT and AST and induced hepatic lipid accumulation. Proteomic analysis revealed that key lipogenic enzymes (Scd1, Fasn, Acly, Acaca) were uniformly suppressed rather than upregulated. Proteins essential for VLDL assembly (ApoB, MTTP, ApoE) were significantly downregulated. Plasma triglycerides were decreased while liver triglycerides were increased, consistent with impaired hepatic lipid export. As in renal tubular cells, AKI also disrupted ER protein folding homeostasis in the liver, triggering ER stress. This was evidenced by upregulated levels of the ER chaperone GRP78, increased XBP1 splicing indicative of UPR activation, and elevated expression of the ER stress-induced pro-apoptotic transcription factor CHOP, suggesting that prolonged ER stress may also promote hepatocyte cell death. TLR4/MyD88 signaling was activated, yet inflammatory cytokines were paradoxically reduced, accompanied by Kupffer cell depletion (decreased F4/80) and monocyte infiltration (increased CD68). In 6,996 propensity-matched ICU patients (MIMIC-IV), AKI independently increased the risk of clinically significant liver injury 4-fold (adjusted OR = 4.41). Analysis of 22,727 patients across 208 hospitals (eICU-CRD) identified a lipid dissociation pattern: elevated triglycerides alongside decreased total cholesterol, HDL, and LDL, with dose-dependent scaling across KDIGO stages. Conclusions These data support ER stress-mediated impairment of VLDL export as a primary driver of AKI-induced hepatic steatosis. Clinical validation across two independent ICU databases identifies a dual metabolic insult: enhanced peripheral lipid delivery compounds impaired hepatic export, amplifying hepatic lipid retention. ER stress and lipid export machinery represent potential therapeutic targets for AKI-associated liver injury.

Abstract INTRODUCTION To examine the generalizability of Alzheimer's disease (AD) biomarker models in real‐world older adults, we examined AD biomarker relationships with cognition in two multicenter cohorts that differ with respect to recruitment approach and health risk factors but were matched on a variety of characteristics. METHODS We compared harmonized health and demographic data, AD and cerebrovascular biomarkers, and cognitive performance in the community‐recruited U.S. Study to Protect Brain Health Through Lifestyle Intervention to Reduce Risk (U.S. POINTER) Imaging substudy and a matched sample from the Alzheimer's Disease Neuroimaging Initiative (ADNI) which recruited primarily from academic specialty clinics. RESULTS Elevated β‐amyloid (Aβ) and tau were associated with cognitive performance in ADNI but not U.S. POINTER. Findings were consistent across different cohort matching schemes, and were not explained by discrepancies in vascular risk. DISCUSSION The role of Aβ and tau in cognitive performance may be reduced in real world samples compared to academic specialty clinics.

Abstract Background Single breath diffusing capacity of the lung for carbon monoxide (DLCO) remains the most widely used test to assess pulmonary gas exchange. GLI DLCO reference equations are based on a larger, more diverse dataset and offer age-specific and internally consistent reference values. However, their impact on interpretation in a real-world population remains uncertain. Research Question Describe the direction and magnitude of DLCO interpretation changes resulting from our multicenter PFT laboratory transition from Miller to GLI reference equations. Study Design and Methods In this multicenter cross-sectional study, we reanalyzed 88,060 DLCO measurements in a large pulmonary function laboratory database performed between 2016 and 2022 using GLI and Miller reference equations, stratified by self-reported race and ethnicity. Baseline patient characteristics and lung function measurements were summarized with continuous characteristics as median (IQR) and categorical variables as n (%). Kruskal-Wallis test was used to evaluate the difference between GLI and Miller DLCO % predicted difference between races, and two-by-two and four-by-four tables were generated to summarize the proportion of concordance and discordance of GLI to Miller severity levels. A repeated measures multivariable logistic regression was used to evaluate discordance between GLI and Miller severity and patient characteristics (age, sex, race, and BMI). Results We included 63,357 individuals with 88,060 PFT visits; 91.6% (58,076) identified as Caucasian, 5.2% (3,262) as Black, 1.4 % (895) as Asian, and 1.8% (1,113) as mixed/other race. Fifty-one percent were female with an median age of 64 years (range: 5- 85). 7.4% (6509) of DLCO results were discordant, with 0.5% (n = 439) abnormal by GLI only, and 6.9% (n = 6070) abnormal by Miller only. Compared to the Miller, GLI interpretations were associated with a greater shift toward normal interpretations. Younger age, lower BMI, female sex, and being of NE Asian, SE Asian, or Other/mixed race compared to Caucasian were associated with higher odds of discordance in DLCO severity/normal classification. Interpretation The use of GLI reference equations substantially alters the interpretation of DLCO results, shifting a meaningful subset of patients from abnormal to normal classification. While GLI reference equations offer more comprehensive, population-based approach, it is crucial to place these interpretations within the context of the patient presentation until further investigations more clearly define the clinical implications of these transitions. This abstract is funded by: none