Citations

e18026 Background: Immune checkpoint inhibitor (ICI) therapy is part of standard of care first-line therapy for recurrent or metastatic (R/M) HNSCC. However, less than 20% of patients (CPS ≥1) have an objective response to ICI monotherapy, and most tumors progress within ~3 months. We sought to improve the proportion of patients who benefit from a chemotherapy-free regimen by combining intratumoral (IT) VV1 oncolytic virus (VSV-IFNβ-NIS) injections with the anti-PD-1 antibody, cemiplimab. We report a pre-planned analysis of Simon's stage 1 results. Methods: We activated a phase 2 study testing the efficacy of first line IT VV1 plus cemiplimab in 1 st line patients with R/M HNSCC. A Simon's two-stage design was employed whereby 10 patients would be accrued in stage 1, and if 3 or more responses were observed, 12 additional patients would be accrued for stage 2. The primary endpoint was objective response rate (ORR) per investigator (RECIST 1.1). Eligible patients had untreated R/M HNSCC from primary oropharynx, oral cavity, hypopharynx, or larynx sites with CPS ≥1. Patients had to have at least one measurable lesion amenable to IT injection. Treatment comprised of IT injections to 1-5 lesions of oncolytic VV-1 virus and IV cemiplimab 350 mg on day 1, repeated every 21 days. Each lesion could be injected 3 times; a 4th injection was allowed if a partial response (PR) was achieved after 3 injections. Cemiplimab could be continued for up to 2 years until progression or unacceptable toxicity. Results: Patient and tumor characteristics, as well as efficacy and toxicity are summarized in the table below. Ten patients were treated at the time of Stage 1 analysis, with 3 showing a confirmed PR for ORR of 30%. Responder CPS scores were 45, 12 and 20%. Two of the responders subsequently showed progression after 6.2 and 9.9 months, the third has maintained PR at 7.7 months to date. Eight of 10 patients showed decrease in the sum of target lesion size (mean decrease of 16%; range -57 to 77%). One grade 2 event of ICI nephritis resolved with corticosteroid treatment. Conclusions: Three patients showed confirmed PR by RECIST 1.1 in Simon's stage 1, meeting criteria to proceed to Simon's stage 2. A total of 22 patients in the HNSCC will be enrolled. VV1 plus cemiplimab shows promise in first line R/M HNSCC based on these initial results. Clinical trial information: NCT04291105 . Patient demographics. N (treated) 10 Sex 20% female Median age (y) 60 (54 – 76) Oropharynx 7 HPV+ 7 Oral cavity 3 PD-L1 CPS median (range) 7.5 (1-45%) Locoregional injection 3 Distant injection 6 Local and distant injection 1 Best response (RECIST 1.1) PR 3 SD 5 PD 2 Adverse Events #pts any VV1/cemiplimab-related G3 AE 4 Lymphopenia 3 Elevated ALK 1

e17594 Background: Cardiovascular disease (CVD) is the leading cause of death among women with ovarian cancer. This study investigates cardiovascular mortality in women with ovarian cancer across the United States, assessing trends by geography and urbanization. Methods: Mortality data from the CDC WONDER database (1999–2020) were analyzed. Females with ovarian cancer listed on the death certificate and cardiovascular disease as the cause of death were identified using ICD-10 codes. Age-adjusted mortality rates (per 100,000) were calculated and stratified by census region and urbanization. Trends were evaluated using Average Annual Percent Change (AAPC) calculated by the Joinpoint Regression Program V5.4.0. Results: From 1999 to 2018, cardiovascular mortality in ovarian cancer patients declined, with an APC of –2.09%, consistent across regions and urbanization levels. However, from 2018 to 2020, a concerning reversal occurred, with mortality increasing by +3.79% per year. The abrupt rise in mortality is observed in the Midwest region, with an APC of +3.99, and in small and medium metro regions, with APCs of +7.38 and +6.46, respectively. This reversal in cardiovascular mortality in ovarian cancer from 2018 to 2020 is likely due to the COVID-19 pandemic that disrupted screening and treatment. Conclusions: This 21-year analysis demonstrates a long-term decline in cardiovascular mortality among women with ovarian cancer, followed by a sharp reversal from 2018 to 2020. This increase suggests the urgent need for public health interventions to reduce cardiovascular risk among ovarian cancer patients.

e17524 Background: Cervical cancer remains one of the leading causes of female cancer mortality, while heart failure drives chronic disease burden. Their coexistence complicates outcomes through shared risk factors, cardiotoxicity, and mortality. This study examined cervical cancer mortality trends in patients with heart failure, assessed disparities, and aimed to inform integrated management strategies. Methods: We analyzed death certificates from CDC WONDER Database (1999–2020), using ICD-10 C53 (cervical cancer) and I50 (heart failure). Female mortality was stratified by race and urbanization. Joinpoint Regression Program V5.4.0 calculated Annual Percent Change (APC) and Average Annual Percent Change (AAPC). Results: A total of 5,534 deaths were identified. Age-adjusted mortality rates (AAMR) declined overall, but trends varied. Among females, mortality fell from 0.1102 per 100,000 in 1999 to 0.076 in 2020 (AAPC -1.28%). Three phases emerged: a modest decline (1999-2006, APC -1.45%), a rapid decline (2006-2009, APC -17.6%), and a concerning reversal (2009-2020, APC +3.27%). The latter increase may reflect reduced screening, declining vaccination rates, or COVID-19 disruptions.By urbanization, large central metropolitan areas showed a steady decline (APC -4.13%). By race, White women demonstrated a consistent reduction (APC -2.61%), indicating sustained benefits of screening and vaccination. Conclusions: Cervical cancer mortality in heart failure patients declined from 1999-2020, reflecting prevention and healthcare advances. However, disparities persist, with notable increases among women since 2009. These findings highlight successes but stress the urgent need for vaccination and screening.

e17134 Background: Despite being the most prevalent malignancy among men, prostate cancer (PC) lacks FDA-approved primary prevention therapies. Current options, such as 5-alpha reductase inhibitors (5-ARIs), offer only modest risk reduction—approximately 23–25%—as shown in the PCPT and REDUCE trials. Emerging preclinical data indicate that GLP-1 receptor agonists (GLP-1RAs) may offer a novel alternative by inhibiting oncogenic pathways and inducing apoptosis in PC cell lines. To investigate this potential, we conducted the first real-world comparative analysis of PC incidence among high-risk men receiving GLP-1RAs versus 5-ARIs. Methods: Using the TriNetX global collaborative network (150M patients), we identified men with elevated PC risk (genetic risk, family history, or high-risk demographics/comorbidities). GLP-1RA users (cohort A) were propensity-score-matched 1:1 with 5-ARI users (cohort B) on demographics, metabolic comorbidities, baseline PSA, BMI and HbA1c. A minimum of 2 month’s prescription was required. The primary endpoint was incident PC. The secondary endpoint was adverse events related to GLP-1RA and 5-ARI. Patients with outcomes before the study inclusion were excluded. Sensitivity and subgroup analyses were performed, adjusting for screening frequency and baseline BMI. Kaplan–Meier curves assessed cumulative incidence, Cox models estimated hazard ratios (HRs, 95% CI), and logistic regression analyzed binary outcomes. Results: After matching, 118,904 high-risk men were identified (59,452 patients per cohort). Cohorts had comparable demographics with a mean age of 48 years, 77% White, 9% Black, and 2% Asian. GLP-1RA users had higher mean A1C (6.3% vs 5.7%) and BMI (38 vs 31), while 5-ARI users had higher mean PSA (3.1 vs 1.3). Median follow-up was 7 years for both cohorts. PC incidence was 2.10% (1,245/59,189) in GLP-1RAs users versus 2.85% (1,689/59,164) in 5-ARI users (NNT 133). GLP-1RAs were associated with a 20% reduction in PC risk (HR: 0.803; CI: 0.746–0.864). Safety outcomes showed that GLP-1RAs were associated with an increased risk of erectile dysfunction and nausea/vomiting. In contrast, GLP-1RAs were associated with lower risks of abdominal pain and constipation. No significant associations were observed for gynecomastia or diarrhea. Conclusions: In this large-scale, head-to-head comparison, GLP-1RAs demonstrated superior efficacy over 5-ARIs, providing a 20% relative risk reduction in PC incidence among high-risk men. Given the global surge in GLP-1RA utilization, these findings suggest a profound public health opportunity: the ability to achieve systemic oncologic protection through existing metabolic therapies. If validated by prospective trials, GLP-1RAs could transition from metabolic blockbusters to a cornerstone of precision oncology and primary cancer prevention.

e17104 Background: Prostate cancer is a commonly diagnosed malignancy among men. Sepsis continues to pose a significant threat to outcomes. Patients with are particularly vulnerable to sepsis due to immunosuppression from the malignancy itself, advanced age, and treatment-related factors including catheter use. Sepsis is associated with high short-term mortality, and worsened overall outcomes, underscoring the need for epidemiological and clinical evaluation. Methods: We analyzed death certificates from 1990-2023 from CDC Wonder Database. We used ICD-10 codes, C61 for malignant neoplasm of prostate and C56 for sepsis to identify related deaths. Mortality rates were compared by urbanization, sex and census region. Joinpoint Regression Program V5.4.0 computed Annual Percent Change with 95% CIs. p value < 0.05 was considered statistically significant. Results: Between 1999 and 2023, 39,801 deaths were recorded. Overall age-adjusted mortality rates (AAMRs) changed little over time, remaining near 2 deaths per 100,000 population (2.02 in 1999 and 2.00 in the most recent year). Clear differences were seen across demographic groups.In 1999, AAMRs among Black or African American was highest with value of 6.80 compared to white and hispanic or latino individual with AAMR vales 1.67 and 1.89 respectively.In 2023, mortality among Black individuals fell to 5.06; however, rates were still higher than those seen among White (1.76) and Hispanic/Latino (1.92) populations. Geographic differences were also evident.The mortality rate in the South, decreasing from 2.43 in 1999 to 2.16 in 2023, was highest among all the other census regions.Mortality rates declined across both metropolitan and nonmetropolitan areas, falling from 2.02 to 1.84 in metropolitan regions and from 2.05 to 1.73 in nonmetropolitan regions in year 2020. Conclusions: Overall sepsis-related mortality has remained relatively unchanged over the two decades, with significant racial and geographic disparities. Black individuals continue to experience disproportionately higher mortality, and Southern region showed consistently elevated rates. Both metropolitan and non-metropolitan areas saw modest declines. These findings highlight the continued need for targeted preventive measures and management strategies to reduce mortality and improve outcomes. Prostate cancer with sepsis AAMRs. variable AAMR 1999 AAMR 2023 Overall / Male 2.02 (1.91-2.14) 2.00 (1.92-2.09) Race Black or African American 6.80 (6.04-7.56) 5.06 (4.60-5.53) White 1.67 (1.56-1.78) 1.76 (1.67-1.85) Hispanic or Latino 1.89 (1.39-2.50) 1.92 (1.65-2.20) Census Region Northeast 2.25 (2.00-2.51) 1.99 (1.80-2.18) Midwest 1.71 (1.50-1.92) 1.65 (1.49-1.81) South 2.43 (2.22-2.63) 2.16 (2.02-2.30) West 1.54 (1.33-1.76) 2.12 (1.94-2.30) Urbanization AAMR 2020 Metro 2.02 (1.90-2.15) 1.84 (1.75-1.93) Non-metro 2.05 (1.79-2.30) 1.73 (1.54-1.92)

e17027 Background: Guidelines recommend active surveillance for low-risk prostate cancer and definitive therapy for high-risk disease. We evaluated national patterns of risk-misaligned care and the survival consequences of undertreatment. Methods: National Cancer Database prostate cancer cases diagnosed from 2010–2022 were analyzed. Low-risk disease was defined as clinical T1–T2, Grade Group 1, and PSA < 10; overtreatment was definitive treatment vs active surveillance or no treatment. High-risk disease was defined as Grade Group ≥4, PSA ≥20, or clinical T3–T4; undertreatment was no definitive local therapy or incomplete definitive therapy. Multivariable logistic regression evaluated factors associated with misalignment. Overall survival (OS) among high-risk patients was assessed using multivariable Cox regression. Results: Among low-risk patients (N = 124,562), overtreatment occurred in 40% of cases. Overtreatment was more common among young men (adjusted odds ratio [aOR] 0.982; p < 0.001) and less common at academic centers (aOR 0.414; p = 0.015). Compared with White men, odds of overtreatment were lower among Black men (aOR 0.946; p = 0.002) and Asian men (aOR 0.827; p < 0.001), with additional variation by education and region (all p < 0.001). Among high-risk patients (N = 318,532), undertreatment affected 27.3% of patients and was more common among older men (aOR 1.061; p < 0.001). Compared with White men, undertreatment was higher among Black men (aOR 1.479; p < 0.001) and lower among Asian patients (aOR 0.863; p < 0.001) (overall race p < 0.001). Compared with private insurance, undertreatment was higher in Medicaid (aOR 2.038; p < 0.001) and uninsured patients (aOR 2.649; p < 0.001), and slightly lower with Medicare (aOR 0.931; p < 0.001) (overall insurance p < 0.001). Multiple comorbidities were associated with higher undertreatment (≥2 vs 0: aOR 1.252; p < 0.001). Facility type was not independently associated (p = 0.490), while region remained significant (overall p < 0.001). In the survival cohort (high risk) (N = 338,919), undertreatment was associated with substantially worse OS (HR 2.745, 95% CI 2.706–2.784; p < 0.001), adjusted for age (HR 1.056; p < 0.001), race (p = 0.001), insurance (p < 0.001), comorbidity (p < 0.001), income (overall p < 0.001), education (overall p < 0.001), and facility region (overall p < 0.001). Conclusions: Low-risk overtreatment remains common nationwide and varies across patient and health-system factors. In high-risk disease, undertreatment affects over one-quarter of patients and disproportionately impacts Black men and those with Medicaid or no insurance. Undertreatment confers a nearly threefold mortality hazard (adjusted HR 2.75; p < 0.001), highlighting a major national quality gap and the urgent need for system-level interventions to ensure equitable, guideline-concordant delivery of definitive therapy.

e17005 Background: We previously developed a hybrid framework combining LLMs and rule-based algorithms (RBA) for automated risk stratification in localized prostate cancer (locPCa), achieving 89% accuracy. However, clinical T-stage (cTS) inference from narrative MRI reports remained the primary source of errors. Herein, we evaluated whether ensemble approaches using multiple state-of-the-art LLMs could improve cTS inference and overall NCCN risk stratification accuracy. Methods: This study included patients with locPCa presenting with at least one positive prostate biopsy and MRI report available. Building on our previously validated hybrid framework that extracts structured phenotypic variables (PSA, prostate volume, cTS, Gleason patterns, grade group, positive and examined cores) from unstructured reports using LLMs and applies NCCN-concordant RBA for risk stratification, we focused on improving cTS inference. Four HIPAA-compliant LLMs (GPT-4.1, GPT-5.1, Gemini-2.5-Pro, Gemini-2.5-Flash) were evaluated using structured zero-shot prompts iteratively refined through oncologist feedback. Majority voting was employed to generate consensus cTS predictions. Performance was evaluated using weighted accuracy. Results: A total of 358 patients were included. The median age at diagnosis was 65 years (IQR: 60-69); the majority were White (92%) and non-Hispanic (94%). cTS distribution was T2a (51%), T2c (20%), T3a (16%), T3b (7%), T2b (4%), and T4 (2%). The most prevalent risk category was unfavorable intermediate (38%) followed by favorable intermediate (20%), high risk (19%), very high risk (16%), low (5%), and very low (1%). Individual LLM accuracy for cTS inference ranged from 85-90%, with Gemini-2.5-Pro achieving highest accuracy (90%), followed by GPT-4.1 (88%), GPT-5.1 (87%), and Gemini-2.5-Flash (85%). The four-model ensemble reached consensus in 92% of cases, achieving 93% cTS inference accuracy. When integrated into NCCN risk stratification schema, single-model approaches achieved 89-92% accuracy, while ensemble consensus cases achieved 95% accuracy. Conclusions: The ensemble framework using multiple LLMs significantly improves cTS inference and achieves 95% NCCN risk stratification accuracy in consensus cases (92% of patients), representing a 6-percentage-point improvement over our previous single-model framework (89%). This strategy demonstrates that ensemble methods can overcome individual model limitations for complex clinical tasks and holds promise for reliable AI-assisted risk stratification.

e17004 Background: Living clinical practice guidelines offer the opportunity to keep recommendations aligned with the latest evidence, but they also present a fundamental challenge: as evidence accumulates, panelists cannot sustainably query and synthesize a growing body of literature. In developing ASCO's living metastatic castration-resistant prostate cancer (mCRPC) guideline, we encountered this challenge firsthand. To address it, we developed and evaluated an interactive agentic framework that enables guideline panelists to query evidence in natural language and retrieve structured facts, tables, and figures—with the goal of extending this tool as a clinician-facing companion to enhance guideline accessibility. Methods: The mCRPC systematic review dataset supporting ASCO's clinical practice guideline was used, comprising 212 structured columns capturing trial characteristics, patient populations, and outcomes extracted from 188 references across 88 clinical trials. A large language model (LLM)-based agentic framework was developed employing a coding agent that translates natural language queries into executable scripts for structured evidence retrieval. Claude (Haiku 4.5, Sonnet 4.5, and Opus 4.5) served as the core LLM agents. We constructed a benchmark of 50 fact-based queries with gold-standard answers targeting trial identification, treatment modalities, and endpoint characteristics (e.g., "Which trials evaluated systemic monotherapy?"). Performance was evaluated using exact-match accuracy for final answers, and precision, recall, and F1 score for trial-level retrieval. Results: Across 50 queries, exact-match accuracy ranged from 82% to 86%. Haiku 4.5 and Sonnet 4.5 each achieved 82% (42/50), while Opus 4.5 achieved 86% (43/50). Precision ranged from 0.93–0.97, recall from 0.94–0.96, and F1 scores from 0.91–0.94. Conclusions: This agentic framework enables scalable, auditable, natural language interrogation of guideline evidence, addressing a critical bottleneck in living guideline maintenance. For panelists, it supports rapid, reproducible evidence surveillance during recommendation development. For clinicians, it offers a foundation for an interactive companion tool that transforms static guidelines into queryable resources—enabling personalized evidence retrieval at the point of care. LLM EM Precision Recall F1 claude-haiku-4-5 0.82(42/50) 0.95 0.94 0.91 claude-sonnet-4-5 0.82(42/50) 0.97 0.96 0.93 claude-opus-4-5 0.86(43/50) 0.93 0.96 0.94

e17003 Background: Accurate ascertainment of prostate cancer (PCa) disease states—biochemical recurrence (BCR), castration-sensitive prostate cancer (CSPC), and castration-resistant prostate cancer (CRPC)—is critical for automated clinical trial enrollment, quality improvement initiatives, and real-world evidence generation. Essential information for disease state ascertainment resides in unstructured EHR data, requiring labor-intensive manual abstraction that is inconsistent and not scalable. We developed an automated hybrid AI framework using large language models to identify PCa disease states from real-world EHR data. Methods: This retrospective study included patients with histopathologically confirmed PCa (2017-2022) and ≥3 years follow-up. Sequential EHR data—including PSA and testosterone levels, clinical notes, pathology and radiology reports—were retrieved across temporal points. BCR was identified using rule-based algorithms (post-radical prostatectomy [RP]: PSA ≥0.2 ng/mL ×2; post-radiation therapy [RT]: nadir + 2 ng/mL ×2). CRPC was determined using androgen deprivation therapy (ADT) timing, PSA kinetics, testosterone < 50 ng/dL, and radiographic progression per PCWG/EAU-ASCO criteria. GPT-4o with structured parametrized prompts extracted metastatic status and radiographic progression. Independent clinicians performed manual annotation for validation. The framework was developed on 10% of data with evaluation on 90%. External validation used a held-out cohort of clinical trial patients. Results: Among 200 PCa patients (median age 66 years [IQR 60-71]; 91% White, 93% non-Hispanic), the framework demonstrated high accuracy: PCa diagnosis (92%), RP dates (94%), RT dates (93%), and ADT initiation (92%). For disease state identification, BCR detection achieved 90% accuracy, castration status 93%, and metastatic state (M0/M1) 87%. External validation on trial-enrolled patients confirmed robust performance: 26/31 (84%) CRPC trial patients and 26/27 (96%) CSPC trial patients were correctly classified, validating the framework’s reliability for clinical trial matching. Conclusions: This hybrid LLM-based framework accurately identifies PCa disease states using real-world EHR data with minimal manual intervention. The system offers a scalable, reproducible approach for automated disease state identification, enabling streamlined clinical trial matching and enhanced real-world data curation in oncology.

e16598 Background: Brain metastases occur in a substantial proportion of patients with renal cell carcinoma (RCC) and are associated with poor prognosis. Stereotactic radiosurgery (SRS) is frequently used for intracranial disease control; however, contemporary real-world survival outcomes and the influence of systemic therapy exposure and metastatic burden remain incompletely characterized. Methods: We conducted a retrospective observational study using the TriNetX global federated health research network. Adult patients (≥18 years) with RCC and documented brain metastases who underwent stereotactic radiosurgery between January 2013 and December 2023 were identified. The index date was defined as the date of first SRS. Baseline demographics, comorbidities, and prior systemic therapy exposure including immune checkpoint inhibitors and VEGF-targeted therapies within 6 months before SRS were assessed. Metastatic burden was approximated using the number of distinct extracranial metastatic organ sites (single-organ vs multi-organ involvement). Overall survival (OS) was evaluated using Kaplan–Meier methods. Results: A total of 232 patients with RCC and brain metastases treated with SRS were included. Mean age was 61.2 ± 10.6 years, and 72% were male. Common comorbidities included hypertension (53%), diabetes mellitus (30%), chronic kidney disease (28%), and cerebrovascular disease (26%). Prior to SRS, 78% of patients received systemic therapy, including immune checkpoint inhibitors (49%) and VEGF-targeted therapies (46%). Using organ-site count as a proxy for disease burden, 59% demonstrated single-organ extracranial metastatic involvement. During follow-up, 95 patients (40.9%) experienced all-cause mortality. Median overall survival was 967 days (31.8 months). Estimated long-term survival probability at maximum follow-up was 21.2%. Kaplan–Meier analysis demonstrated early survival attrition followed by prolonged survival in a subset of patients. Patients receiving immunotherapy and those with limited extracranial metastatic burden demonstrated more favorable survival patterns. Conclusions: In this large contemporary real-world cohort of RCC patients with brain metastases treated with stereotactic radiosurgery, median overall survival exceeded 2.5 years, providing important modern survival benchmarks. Favorable outcomes were observed among patients receiving systemic immunotherapy and those with limited metastatic burden, supporting the continued role of SRS within multidisciplinary management strategies for selected patients with metastatic RCC.