Citations

TPS11598 Background: Fibroblast Activation Protein (FAP) is expressed on the surface of cancer-associated fibroblasts and is involved in the formation, maintenance, and spread of tumors. Sarcoma frequently demonstrates high expression of FAP on the tumor cells themselves in contrast to other solid tumors, where it is mainly expressed on CAFs within the tumor stroma. Radioligand therapies utilizing alpha emitting isotopes such as Ac-225 have emerged as promising treatment strategies across multiple targets and indications in oncology. Ac-225 and its daughter radionuclides emit high-energy alpha particles capable of inducing double-stranded DNA breaks, leading to significant cytotoxic effects within cells. Early clinical data for other Ac-225-labeled compounds have shown promising outcomes, particularly in settings where traditional therapies have failed. [Ac-225] RTX-2358 (Ratio Therapeutics Inc, Boston, MA) is a novel FAP-targeted radioligand designed for specific high-affinity binding, resulting in prolonged tumor residence and limited normal tissue uptake. This sustained tumor retention leads to extended delivery of cytotoxic radiation to FAP-expressing tumors. A beta-emitting analog, [Lu-177]RTX-2358, has been administered to more than 25 patients under the German §13.2b compassionate use framework across multiple tumor types. Longitudinal SPECT imaging performed as late as 14 days after administration demonstrated persistent tumor retention with minimal washout. Based on time–activity curve analysis from these data, tumor biological half-time is estimated to be well in excess of 1,000 hours. Renal clearance was observed to be consistent with that reported for established lutetium-177–labeled radioligand therapies. Methods: This study is a seamless Phase 1 / 2 dose escalation study. The Phase 1 portion consists of a 3 x 3 dose escalation design, with patients treated in 3 planned dose escalation cohorts (7.5, 11.25 and 15 MBq) with up to 6, 8-week cycles of [Ac-225]RTX-2358. Backfill of phase 1 cohorts is permitted to facilitate selection of the recommended administered activity level for the subsequent Phase 2 expansion cohort. Dose escalation / de-escalation decisions will be made by a Safety Review Committee. Patients over 18yrs with a history of histologically confirmed relapsed or refractory soft tissue sarcoma (measurable disease per RECIST v1.1) are eligible if they have a positive [Cu-64]LNTH-1363S FAP PET, an ECOG PS 0 to 1, and adequate organ reserve and renal function. Recruitment of the first cohort was completed in December 2025. The ATLAS Study: NCT07156565 Clinical trial information: NCT07156565 .

e22677 Background: Lung cancer without tobacco exposure is a growing and preventable source of global morbidity driven by environmental and occupational exposures, with marked regional variation and unevenly addressed prevention strategies. Methods: Global Burden of Disease (GBD) 2023 estimates were used to quantify lung cancer disability-adjusted life years (DALYs) and age-standardized mortality rates (ASMRs) attributable to ambient particulate matter (PM2.5), household air pollution (HAP), residential radon, and occupational carcinogens across seven GBD super-regions from 1990–2023. Population attributable fractions (PAFs) and temporal trends were assessed, with sex-stratified analyses for radon and occupational exposures. Results: Air pollution accounted for the largest attributable lung cancer DALYs in low and middle-income regions, particularly South Asia, where PM2.5 (PAF 15.9%) and HAP (PAF 16.8%) together accounted for 32.7% of DALYs. In Sub-Saharan Africa, HAP alone accounted for 21.4%. Occupational carcinogens contributed the largest proportion in high-income regions (PAF 21.7%) and were associated with the highest DALY rate (230/100,000) and ASMR (13/100,000). From 1990–2023, HAP-attributable lung cancer burden declined across all regions (40–70%), consistent with transitions away from solid household fuels. In contrast, PM2.5-attributable DALY rates declined by 30–40% in high-income regions but increased by 40–50% in South Asia and the Middle East and North Africa. Radon- and occupational carcinogen–attributable burdens remained largely stable across most regions, although high-income regions experienced declines in radon (18.4%) and occupational carcinogen (32.1%) attributable DALY rates. Radon showed a modest male excess (male-to-female ratios 1.2–1.5), whereas occupational carcinogens demonstrated marked male predominance (often > 4). Conclusions: Despite substantial decrease in HAP related lung cancer burden, other modifiable factors including PM2.5, occupational carcinogen and radon exposure remain major contributors, specially in high-income regions where occupational exposures account for the largest PAF. These patterns suggest that reducing household energy exposure alone is insufficient. We need regional exposure targeted prevention strategies that extend beyond tobacco control and are relevant to oncology practice, survivorship and policy advocacy. Population-attributable fractions of lung cancer burden from environmental and occupational exposures across gbd super-regions. GBD Super-Region Air Pollution (PM2.5 + HAP) Residential Radon Occupational Carcinogens High-income 6.2 4.6 21.7 Central Europe, Eastern Europe & Central Asia 11.3 6.3 10.9 Latin America & Caribbean 12.5 4.1 11.9 North Africa & Middle East 21 3.7 12.7 South Asia 32.7 4.3 10 Sub-Saharan Africa 30.7 4.3 9.4 Southeast Asia, East Asia & Oceania 23.6 3.4 9.8

e20511 Background: Trop-2 ADCs have emerged as a novel therapeutic approach in NSCLC that is refractory to chemotherapy and immune checkpoint inhibitors. Comparative interpretation of the efficacy of multiple Trop-2 ADCs, including Sacituzumab tirumotecan (Sac-TMT), Datopotamab deruxtecan (Dato-DXd), and Sacituzumab govitecan (SG), is limited. We therefore conducted a systematic review, IPD survival analysis, and meta-analysis to evaluate the efficacy and PROs of Trop-2 ADCs compared with chemotherapy in NSCLC. Methods: A systematic search was conducted from inception to October 31, 2025, across PubMed, Scopus, Cochrane, and CT.gov. Clinical trials evaluating Overall Survival (OS) and Progression-free survival (PFS) in patients treated with Trop-2 ADC alone or compared with Chemotherapy were included. IPD were extracted from the Kaplan-Meier curves of clinical trials for reconstruction using the KMfromIPD website by MD Anderson, which stratified the data by treatment group. Outcomes included OS, PFS, and time to deterioration (TTD). The cumulative data were analyzed using R 4.5.1 and RStudio. RevMan version 5.4.1 was used for the meta-analysis of subgroups. Heterogeneity was assessed using the I 2 statistic. Results: Of 513 articles, 7 phase 1/2 single-arm and 4 randomized control trials were included, comprising 2,619 patients.Trop-2 ADCs were associated with improved OS (Median 15.6 vs 12.5 months; HR 0.78, 95% CI 0.69-0.88; p < 0.0001) and PFS (Median 5.5 vs 4.1 months; HR 0.66, 95% CI 0.59-0.74; p < 0.0001). Sac-TMT demonstrated the greatest OS benefit (OS 23.3 months; HR 0.40). Further drug classes are depicted in Table 1. PROs assessment showed TROP-2 ADCs were associated with longer time to deterioration in global quality of life and dyspnea-free survival compared to chemotherapy (HR 0.8 and 0.69), (p = 0.04 and p < 0.0001, respectively). Exploratory subgroup analysis showed improved OS among patients with actionable gene alteration (AGA) treated with Trop-2 ADCs (HR 0.58, p < 0.00001). However, no significant difference was observed in either treatment arm among patients without AGA (HR 0.89, p = 0.15). Conclusions: Trop-2 ADCs were associated with improved survival and PROs compared with chemotherapy in NSCLC, with variability across agents. Sac-TMT has shown higher OS and PFS. Dato Dxd had a favorable PFS; however, the OS was not superior to chemo. Furthermore, SG showed no favorable clinical outcomes compared with the chemo group. These findings support the continued development of Trop-2-directed therapies and highlight the need for prospective comparative trials and biomarker-driven patient selection. Drug N mOS (months) HR p- value mPFS (months) HR p- value Sac-TMT 386 23.32 0.4 <0.000001 7.5 0.5 <0.000001 Dato-DXd 486 13.2 0.95 0.55 5.02 0.71 <0.000001 SG 347 12.61 1.02 0.82 4.18 0.87 0.05

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.

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.

e15713 Background: Fruquintinib demonstrated overall survival benefit in heavily pretreated metastatic colorectal cancer (mCRC) in FRESCO and FRESCO-2. The FDA approved fruquintinib on Nov 8, 2023 for adult patients with mCRC after standard chemotherapy plus anti-VEGF therapy, with anti-EGFR therapy when appropriate (e.g., RAS wild-type). Observed geographic variation in post-approval uptake suggests potential disparities in the delivery of evidence-based medicine across U.S. regions. Methods: We conducted a retrospective cohort study using Epic Cosmos, a database of over 300 million patient records from over 1,500 hospitals nationwide. We identified adults with metastatic colorectal cancer (mCRC) with prior standard systemic therapy exposure and recorded fruquintinib exposure following FDA approval beginning Nov 8, 2023. Fruquintinib uptake was analyzed by U.S. Census region and state, expressed as the proportion of eligible mCRC patients receiving fruquintinib within each geography. Results: Among 116,116 pre-treated mCRC patients, 6,722 (5.79%) had recorded fruquintinib exposure during the post-approval period. Uptake varied nearly five-fold across states (2.27%–11.33%) (Table). High-uptake states were disproportionately concentrated in the central U.S./Midwest, whereas multiple low-uptake states were observed in the West and Northeast, suggesting broader regional variation beyond state-to-state differences. Conclusions: In a large national EHR-derived cohort of pre-treated mCRC patients after U.S. approval, fruquintinib uptake demonstrated substantial geographic variation. These findings raise concern for inequities in delivery of evidence-based later-line therapy and highlight the need to identify modifiable system- and site-level drivers (e.g., access, referral patterns, treatment infrastructure) to promote equitable adoption. Fruquintinib uptake among eligible metastatic colorectal cancer patients by U.S. state. Rank Geography Eligible mCRC (N) Fruquintinib uptake (%) Overall Overall 116,116 5.79 Top 1 Oklahoma 1,236 11.33 2 Kansas 1,312 10.75 3 Indiana 2,448 10.70 4 Missouri 1,521 9.47 5 Connecticut 2,208 9.24 Bottom 1 Washington 1,232 2.27 2 Nevada 441 2.72 3 Rhode Island 476 2.73 4 Tennessee 1490 2.82 5 Maine 614 2.93 Fruquintinib uptake is reported as the proportion of eligible pre-treated metastatic colorectal cancer (mCRC) patients with recorded fruquintinib exposure following U.S. FDA approval. States are ranked by uptake percentage.

e14524 Background: Immune checkpoint inhibition has demonstrated meaningful clinical benefit in microsatellite instability–high (MSI-H) colorectal cancer (CRC); however, outcomes remain heterogeneous. The interaction between tumor immune microenvironment features, tumor mutational burden (TMB), and survival outcomes has not been fully characterized. We conducted an integrative analysis of publicly available genomic data to evaluate the prognostic impact of immune infiltration stratified by MSI status and TMB in CRC. Methods: Transcriptomic, somatic mutation, and clinical data for primary colorectal adenocarcinoma were obtained from TCGA-COAD and TCGA-READ. About 600 patients with overall survival data were analyzed. MSI status was taken from TCGA molecular subtype annotations. TMB was defined as the number of nonsynonymous somatic mutations per tumor and dichotomized at the median.Immune infiltration (CD8⁺ and CD4⁺ T cells, macrophages, neutrophils, dendritic cells) was estimated using TIMER. Patients were classified as high or low infiltration by the median. Overall survival was assessed using Kaplan–Meier analysis and Cox models adjusted for age, sex, and tumor stage., and tumor stage. Results: The cohort included approximately 15% MSI-H tumors, which demonstrated significantly higher TMB compared with microsatellite-stable (MSS) tumors. MSI-H tumors exhibited an inflamed immune phenotype, with approximately twofold higher CD8⁺ T-cell infiltration relative to MSS tumors. Across the overall cohort, high CD8⁺ T-cell infiltration was associated with improved OS compared with low infiltration (median OS not reached vs approximately 60–70 months). In contrast, increased macrophage infiltration was associated with inferior survival outcomes. When stratified by genomic features, the favorable prognostic association of CD8⁺ T-cell infiltration was most pronounced in MSI-H and TMB-high tumors, corresponding to an estimated 30–40% relative reduction in mortality risk compared with MSI-H/TMB-high tumors with low immune infiltration. Immune infiltration demonstrated weaker and less consistent associations with OS in MSS/TMB-low tumors. In multivariable analyses, CD8⁺ T-cell infiltration remained independently associated with improved OS after adjustment for clinical covariates, whereas macrophage infiltration remained independently associated with worse outcomes. Findings were consistent across colon and rectal cancer subgroups. Conclusions: In this integrative analysis of TCGA colorectal cancer datasets, features of the tumor immune microenvironment, together with MSI status and TMB, were associated with survival. These results support immune-genomic stratification and suggest that combining immune and molecular biomarkers may improve prognostic assessment and immunotherapy trial design in colorectal cancer.

e13656 Background: Effective multidisciplinary cancer care is bottlenecked by the cognitive burden of synthesizing fragmented electronic health records across different specialties. Current AI tools often function as isolated chatbots rather than integrated workflow assistants. We developed and validated "OncoSphere AI," an Agentic AI multidisciplinary clinical workflow support tool designed to autonomously orchestrate the multidisciplinary review process by deconstructing complex cases into domain-specific insights. Methods: OncoSphere AI is a novel, multi-agent large language model–driven platform that simulates a multidisciplinary tumor board for cancer care. The architecture uses a parallel reasoning framework in which eight virtual specialty agents—Medical Oncology, Radiation Oncology, Surgical Oncology, Pathology, Radiology, Pharmacy, Clinical Research, and Nurse Navigation—independently analyze structured and unstructured patient data against specialty specific guidelines and contemporary peer-reviewed literature. System performance was assessed via built-in telemetry (task completion, latency, inter-agent agreement) and blinded review by eight expert oncologists using a structured survey to rate data extraction fidelity, guideline concordance, safety, and clinical utility compared with standard manual review. Results: The study cohort (n=20, median age 59.5 years; 55% female) had a median of 2.5 (range 0-7) prior lines of therapy and multiorgan metastatic disease (median: 2, range 1–4), including lung, breast, gastrointestinal (CRC, pancreatic, hepatobiliary), and genitourinary (prostate, renal) malignancies. OncoSphere AI demonstrated high workflow utility (mean rating: 4/5) and strong extraction fidelity, with more than 90% of biomarker extractions rated as a “perfect match” by expert reviewers. Operational efficiency gains were substantial with AI processing averaged 4.7 minutes per case versus 29.3 minutes for manual review, corresponding to an 84% reduction in time (p<0.001). System telemetry revealed adaptive compute behavior, with the platform autonomously allocating 1.73-fold more reasoning tokens to complex, refractory cases than to standard scenarios. Safety review identified only one critical contraindication issue in a heavily pretreated, multi-line setting. Conclusions: OncoSphere AI effectively automates the clinical “perception layer,” consistently transforming heterogeneous longitudinal data into decision-grade insights with high fidelity and marked time savings. By operating as a coordinated ensemble of domain-specialized AI agents rather than a single generic model, it offers a scalable approach to reduce physician cognitive burden and to standardize guideline-concordant care planning in diverse, heavily pretreated oncology populations.

Background/Objectives: Juxtapleural and perifissural nodules are commonly encountered on thoracic imaging, are typically presumed benign, and are often considered to represent intrapulmonary lymph nodes. We evaluated the prevalence, clinical context, and management of juxtapleural nodules over a 10-year period and to assess factors associated with concern for malignancy. Methods: In this retrospective, single-center study, radiology reports from 1 January 2015 to 31 December 2024, were queried using natural language processing to identify examinations containing terminology consistent with juxtapleural nodules. After exclusion of duplicate encounters, 85,435 unique patient encounters were analyzed. Reports were further filtered to identify cases with imaging concern for malignancy and recommendations for follow-up or intervention (N = 659). Clinical characteristics, imaging modality, and prior malignancy history were compared using appropriate statistical tests. Results: Among 85,435 unique encounters, 659 patients (<1%) had juxtapleural nodules considered concerning for malignancy. These patients were older (68 vs. 64 years, p < 0.001), more frequently imaged with PET/CT (8% vs. 5%, p < 0.001), and more likely to be evaluated in inpatient settings (16% vs. 12%, p < 0.001). Of the cohort with a juxtapleural nodule considered concerning for malignancy by imaging criteria, 360 patients (55%) had a prior history of malignancy. In these patients, nodules were more commonly identified in outpatient settings (70% vs. 43%, p < 0.001) and on routine imaging (94% vs. 86%, p < 0.001). Representative cases demonstrated that malignant nodules may mimic benign perifissural morphology on initial imaging. Conclusions: Although the vast majority of reported juxtapleural nodules did not prompt malignancy-directed evaluation, a small subset of cases (approximately 0.8%), particularly in patients with prior malignancy, were interpreted as sufficiently concerning to warrant follow-up imaging or intervention. Because pathologic confirmation and standardized long-term outcome adjudication were not available for all patients, these findings should be interpreted as reflecting real-world radiologist concern and management patterns rather than a definitive malignancy rate.

Abstract Maternal iron deficiency (ID) affects nearly 85% of all pregnancies worldwide by the third trimester and is strongly linked with maternal morbidity, poor birth outcomes, and cognitive and behavioral developmental deficits in children. Despite its burden on maternofetal health, no guidelines exist for routine screening of ID in pregnant individuals in the United States. The World Health Organization uses a diagnostic (ie, nonscreening) ferritin threshold of 15 μg/L for ID in the first trimester of pregnancy, with no guidance for the second and third trimesters. Routine screening for ID in pregnancy could identify a large proportion of mothers who remain undiagnosed and untreated under the current clinical status quo. Accordingly, we conducted, to our knowledge, the first cost-effectiveness analysis of ferritin-based ID screening in pregnant women in the United States, comparing thresholds of (1) 30 μg/L vs (2) 15 μg/L vs (3) no screening from a modified US societal perspective accounting for wages lost to infusion time, infusion administration costs, and the cost of annual hematology follow-up. We found that screening at 30 μg/L was the preferred strategy across extensive base-case, threshold, and sensitivity analyses and across all examined willingness-to-pay thresholds, with an incremental cost-effectiveness ratio of $34 000/quality-adjusted life year (QALY; 95% credible interval, $32 000-$36 200/QALY) compared with no screening. No parameter variation changed this outcome, and screening at 30 μg/L was favored in 100% of 10 000 Monte Carlo iterations. Population-level regular second- and third-trimester screening for ID using physiologically informed ferritin ranges should be considered for all pregnant women in the United States.