St. Luke''s University Health Network

Background:Clinical trial enrollment in oncology remains critically low, with fewer than 5% of eligible adults participating, in large part due to the complexity and labor intensity of eligibility screening. We prospectively evaluated a neuro-symbolic, multi-agent artificial intelligence (AI) platform integrating domain-specific large language model (LLM) agents, an oncology-specific knowledge graph, a real-time recommendation engine, and human-in-the-loop review to determine whether automated extraction and reasoning can safely improve trial identification, efficiency, and equity at scale. Methods:Consecutive patients [N = 3804; Eastern Cooperative Oncology Group (ECOG) 0-2] balanced for cancer type incidence with metastatic or progressive malignancies were screened across a 12-month period. A multi-agent architecture-OncoAgents (LLM-based extraction and reasoning agents), OncoGraph (oncology knowledge graph), OncoRecommend (prioritization engine), and OncoSet (expert-curated corpus)-carried out automated data extraction, harmonization, and trial matching over 157 367 clinical pages (∼86.5 M tokens). Dual oncologists produced a gold standard of trial eligibility labels (Cohen's κ = 0.92). The primary unit of analysis was the patient-trial pair. Baselines included manual screening, GPT-4 zero-shot prompting, GPT-4 chain-of-thought, and frontier GPT-4o extraction/matching benchmarks. Outcomes included sensitivity, specificity, precision, F1 score, calibration of eligibility confidence scores, time-to-recommendation, fairness across demographic subgroups, and operational burden. Results:The multi-agent neuro-symbolic system achieved an F1 score of 0.82 (95% confidence interval 0.81-0.83). In comparison, the GPT-4 zero-shot baseline achieved an F1 of 0.47, and the GPT-4 chain-of-thought baseline achieved an F1 of 0.67. Per-patient screening time decreased from a median of 120 min (manual review) to ∼30 min total (15 min automated processing + 15 min clinical review). Across the cohort, the system processed 157 000 pages, screened 23 912 candidate patient-trial pairs, and produced 17 912 oncologist-confirmed matches, with median time-to-recommendation <7 days. No demographic subgroup exceeded a 10-percentage point F1 gap; the largest observed difference was ∼7 points between white and black/African American patients. Ablation experiments showed that both knowledge graph grounding and multi-agent decomposition contributed materially to performance and efficiency. Eligibility confidence scores exhibited reasonable calibration in the clinically relevant operating range. Conclusions:A neuro-symbolic, multi-agent architecture that couples LLM-based extraction with ontology-grounded, deterministic eligibility reasoning improved the accuracy, throughput, and timeliness of oncology clinical trial matching versus LLM-only baselines, while preserving clinician oversight and maintaining modest subgroup performance gaps. These results support scalable, equity-aware deployment of AI-assisted trial screening in routine oncology practice.

Local failure and leptomeningeal disease (LMD) are both poor outcomes that can occur after resection and post-operative radiosurgery for newly diagnosed brain metastases (BM). There is increasing utilization of collagen-embedded Cesium-131 brachytherapy (GammaTile®) as a method of providing immediate adjuvant radiation therapy. Post-operative LMD rates following GammaTile implantation for newly diagnosed BMs has yet to be reported. The objective was to evaluate the incidence of LMD rates, local control (LC), and survival following resection and GammaTile for newly diagnosed BMs. An ongoing, multicenter, prospective, observational Phase IV non-interventional registry (NCT0442738) was queried to analyze rates of LMD following surgical resection of newly diagnosed BMs. Following resection and GammaTile implantation, we evaluated LMD rates, LC, and overall survival (OS). The Kaplan-Meier method was used to analyze time-to-event outcomes. Fifty-one patients with 55 BMs were analyzed. The median follow-up was 12.4 months. The majority of BMs were in the supratentorial brain (87.3

Hepatic artery dissection means a tear in the wall of the hepatic artery, allowing blood to accumulate between arterial layers, leading to fatal or major sequelae, including ischemia and hepatic injury, pseudoaneurysm, or rupture. Spontaneous hepatic artery dissection is rare, especially in patients without any recent hepatobiliary surgery or trauma. It is a nonspecific and often subtle presentation that can delay diagnosis, leading to major/fatal consequences, including hepatic injury due to ischemia, pseudoaneurysm, and rupture leading to death, which highlights the need for clinical awareness. We report on a 54-year-old man with hypertension, type 1 diabetes mellitus, hyperlipidemia, and a remote motor vehicle accident 12 years prior to our case presentation. He presented with acute epigastric pain. Computed tomography angiography (CTA) demonstrated an isolated dissection of the common and proper hepatic arteries without involvement of the aorta or celiac trunk. He was managed conservatively with anticoagulation and close follow-up. Repeat CTA at one month showed interval remodeling and partial recanalization of the hepatic artery. Although a few cases have been reported, they offer insights into risk factors, clinical presentation, and the range of management strategies. Our report underscores the importance of considering vascular causes in patients with unexplained upper abdominal pain. Early diagnosis and a tailored treatment plan can help prevent serious complications.