Humayun, Muhammad Abdullah

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.

e16502 Background: Accurate postoperative risk stratification in localized clear-cell renal cell carcinoma (ccRCC) informs prognosis, surveillance intensity, and eligibility for adjuvant systemic therapy. Although validated prognostic models (Leibovich 2018, SSIGN, and UISS) exist, they are inconsistently applied in routine practice due to workflow and documentation barriers. Using AI assisted tools within the electronic medical record (EMR) may enable rapid, standardized risk assessment and improve clinical use. Methods: We retrospectively analyzed 92 patients with localized ccRCC undergoing radical nephrectomy and postoperative oncology evaluation (December 2021–September 2025) across three Mayo Clinic sites. Pathology reports and oncology notes were extracted via structured SQL queries. Clinical and pathologic variables were parsed using a zero-shot GPT-4o pipeline. Gold-standard prognostic risk classification and adjuvant therapy eligibility were established by an expert oncology clinician using validated prognostic calculators and KEYNOTE-564 eligibility criteria. AI-generated outputs and novice manual abstraction by two first-year fellows were compared against expert assessment using accuracy, precision, recall, and F1 score. Leibovich and SSIGN were evaluated as three-class models (low/intermediate/high), and UISS as binary risk stratification (high vs non-high). Results: Among 92 patients with localized ccRCC (median age 64 years [range 41–78]), AI-generated outputs demonstrated near-perfect agreement with expert assessment for KEYNOTE-564 adjuvant immunotherapy eligibility (accuracy 97%, precision 100%, recall 97%, F1 score 98%). For three-class prognostic stratification, AI achieved perfect agreement for Leibovich disease-free survival risk (accuracy, precision, recall, and F1 all 100%), outperforming novice abstraction. For SSIGN, AI demonstrated strong multi-class performance (accuracy 91%, precision 91%, recall 93%, F1 score 92%), compared with novice abstraction (accuracy 92%, F1 score 90%). For binary UISS risk stratification, AI achieved 81% accuracy, precision 74%, recall 100%, and F1 score 85% showing a tendency towards false positives. Conclusions: This study demonstrates AI can use existing prognostic and systemic therapy eligibility to accurately risk stratify and identify patients eligible for adjuvant systemic immunotherapy post nephrectomy. For our next steps, we plan to develop a multimodal AI platform integrating pathologic, radiologic, and clinical data to improve prognostication for outcomes and predictive capabilities for patients who would benefit from adjuvant systemic therapy for more personalized medicine.

703 Background: The association of circulating tumor DNA (ctDNA) reduction with treatment response has been investigated across multiple malignancies. This study aims to assess the impact of ctDNA response on survival outcomes in patients receiving enfortumab-pembrolizumab (EV-P). Methods: We identified 32 patients with advanced urothelial carcinoma who received EV-P and had ≥2 ctDNA assessments with ≥1 detectable level. Deep molecular response was defined as >2-log reduction from highest ctDNA detection. Radiographic progression was determined from radiology reports documenting progression by RECIST criteria. Kaplan Meier analyses were performed to compute Hazard Ratios (HR) with 95% confidence intervals (CI) for radiographic progression free survival (rPFS), defined as time from treatment start to radiographic progression or death and overall survival (OS) defined as time from treatment start to death were computed. Log-rank tests were performed to compare PFS and OS in patients with deep molecular response to those without response. A p-value <0.05 indicated a statistically significant association with response. Results: Thirty-two patients were included (median age 71 years; range 39–90). Median follow-up was 10.6 months. At cutoff (10/1/2025), 26 patients were alive and 6 had died. Seventeen (53%) achieved a ≥2-log ctDNA reduction (responders), while 15 (47%) did not (non-responders). Progression-free survival was significantly longer in responders (log-rank P < 0.001); median PFS was 5.9 months (95% CI 2.4–NR) in non-responders and not reached in responders (HR 0.09, 95% CI 0.02–0.44; P = 0.0026). Median overall survival was 12.7 months (95% CI 10.0–NR) in non-responders versus not reached in responders (log-rank P = 0.011; HR 0.11, 95% CI 0.01–0.88; P = 0.037). At cutoff, 94% of responders and 67% of non-responders were alive. Conclusions: Achieving a >2-log decrease in ctDNA during EV + Pembrolizumab therapy was significantly associated with improved progression free and overall survival. These findings support ctDNA kinetics as a potential early biomarker of response warranting prospective validation. Clinical outcomes by >2-log ctDNA reduction status. Endpoint > 2-log drop (n = 17 ≤ 2-log drop (n = 15) Log-rank p Median (mo) ≤ 2-log > 2-log HR (95% CI) P C-index PFS 2 events (12%) 10 events (67%) 0.0002 5.9 (2.4–NA) NR 0.09 (0.02–0.44) 0.0026 0.77 OS 1 event (6%) 6 events (40%) 0.011 12.7 (10.0–NA) NR 0.11 (0.01–0.88) 0.037 0.73

Abstract Background Accurate diagnosis and risk stratification in myeloproliferative neoplasms (MPNs) rely on expert interpretation of complex clinical records. Outside major academic centers, specialist expertise is scarce, and time-consuming chart reviews extend clinic visits and delay treatment decisions, underscoring the need for automated, point-of-care phenotyping for MPN patients. Methods Patients with the ICD codes of PV, ET and MF were identified from the electronic health record. Cases were defined as patients meeting the 2017 WHO diagnostic criteria for PV, ET or MF or those with a definitive MPN diagnosis determined by a hematologist. Controls were the patients who met neither the diagnostic criteria nor had a hematologist-assigned diagnosis. Two clinicians extracted demographic, clinical (symptoms, thrombotic history), laboratory (complete blood count, serum chemistry), genetic (mutations, karyotype), bone marrow morphology, diagnostic mentions and prognostic score information from unstructured clinical notes and pathology reports. Prognostic scores included the conventional risk model for PV, the International Prognostic Score of Thrombosis (IPSET) for ET and the Dynamic International Prognostic Scoring System (DIPSS), DIPSS-plus or Mutation-enhanced International Prognostic Scoring System (MIPSS-70) for MF. Discrepancies in data extraction were resolved by a consensus. A random sample of 510 patients (PV: 85, ET: 85, MF: 85, controls: 255) was split into prompt development (n=60) and test (n=450) sets for iterative workflow engineering and evaluation. The workflow included: (i) a rule-based algorithm for automated retrieval of demographic details, the first bone marrow biopsy report, the first clinical note after biopsy and the latest laboratory results within 90 days preceding biopsy (ii) a HIPAA-compliant LLM (GPT-4 turbo) for data extraction from the clinical notes and pathology reports (iii) a regular expressions-based algorithm for post-processing extracted data into structured tables, (iv) a clinician-informed source prioritization algorithm to select data from preferred sources, (v) a modified WHO-informed diagnostic characterization tool incorporating hematologist-assigned diagnoses to distinguish cases from controls and; (vi) an NCCN guidelines-based algorithm for prognostic tool selection and risk-stratification based on available data. Structured prompts were used to extract demographic, clinical, laboratory, genetic and pathological data in a zero-shot deterministic setting. The performance of automated data extraction, diagnostic characterization and prognostication was compared against the manually curated data. Accuracy and F1 scores were computed for data extraction and prognostication tasks while sensitivity and specificity were calculated to assess diagnostic performance. Results A total of 450 (PV: 150, ET: 150, MF: 150) pathology reports and 50 (PV: 52, ET: 55; MF: 65) clinical notes were identified for 450 test set patients. The test set data extraction accuracy was 95% (reports: 97%, notes: 92%) for PV, 93% (reports: 92%, notes: 94%) for ET and 98% (reports: 99%, notes 97%) for MF. The test set F1 score was 0.95 (reports: 0.96, notes: 0.93) for PV, 0.95 (reports: 0.95, notes: 0.96) for ET and 0.98 (reports: 0.99, notes: 0.97) for MF. PV patients were diagnosed with 100% sensitivity and 98% specificity whereas ET and MF patients were diagnosed with 100% sensitivity and specificity. The accuracy of risk stratification was 100% for PV and ET and 94% for MF. The F1 score for risk stratification was 1.00 for PV and ET and 0.94 for MF. Conclusions The first-in-class automated framework successfully identified pertinent clinical information, performed accurate data extraction from unstructured records and enabled clinician-guided source prioritization for accurate diagnosis and risk stratification of MPN patients. Prospective validation of the framework is required for widespread clinical implementation.

Abstract Background In patients with myelofibrosis (MF) undergoing allogeneic hematopoietic cell transplantation (allo-HCT), graft-versus-host disease (GVHD) prophylaxis has traditionally included a calcineurin inhibitor (CNI) plus methotrexate (MTX). We conducted a retrospective study comparing the safety and efficacy of post-transplant cyclophosphamide (PTCy)-based GVHD prophylaxis to CNI-based regimens. Methods This study was approved by the Mayo Clinic IRB. Patients with myelofibrosis who underwent allo-HCT after 2016 at Mayo Clinic in Arizona and Florida were included. Categorical variables were compared using Chi-square or Fisher's exact tests. Overall survival (OS) was defined from transplant to death, and progression-free survival (PFS) from transplant to relapse or death. OS and PFS were estimated using Kaplan-Meier methods; univariate differences were assessed with log-rank tests. The cumulative incidence of acute and chronic GVHD and time to neutrophil and platelet engraftment were analyzed using competing risk models, with death as a competing event; differences were tested using Gray's test. Cox proportional hazards models were used to estimate adjusted hazard ratios (HRs). Results A total of 110 patients were included. 53 patients received PTCy-based prophylaxis (43 PTCy/Tac/MMF, 1 PTCy/CSA/MMF, 9 PTCy/Siro/MMF), while 57 received non-PTCy regimens (54 Tac/MTX +/- 39 T-cell depletion, 2 CSA/MMF, 1 Tac/MMF). Median age at transplant was 63 (range: PTCy 40-76; others 38-74) years in both groups; 59 were male (58.5% PTCy vs 49.1% others). Median Karnofsky Performance Status (KPS) and HCT–Comorbidity Index (HCT-CI) in patients receiving PTCy vs non-PTCy were 90 vs 80 and 1 vs 3, respectively. Conditioning intensity was similar: 22.6% myeloablative vs 21.7%, and 77.4% reduced-intensity vs 76.7% in PTCy and non-PTCy, respectively (p=1.000). Stem cell source was peripheral blood in all patients except one in the non-PTCy group who received bone marrow. 58 patients had primary MF (54.7% PTCy vs 50.9% others), and 52 had secondary MF (45.3% PTCy vs 49.1% others), including essential thrombocythemia (28.3% PTCy vs 31.6% others), polycythemia vera (17.0% vs 15.8%), and MPN-NOS (0% vs 1.8%). Median follow-up was 26.2 months (95% CI: 20.7-52.7) in the PTCy and 60.6 months (95% CI: 53.1-70.6) in the non-PTCy cohort. CD3 and CD33 donor chimerism ≥95% at day +100 was 66.7% and 86.7% in PTCy vs 63.6% and 92.7% in others, respectively (p=0.50); day +100 driver molecular mutations were present in 32.1% of PTCy and 32.0% of others (p&gt;0.99). Rates of acute GVHD grades ≥2 and ≥3 were not statistically significant, in the PTCy group (37.7% vs 49.1% p=0.229 and 17.0% vs 22.8% p=0.445, respectively). Chronic GVHD (any grade) occurred in 42.3% of PTCy vs 38.2% of others (p=0.664), while moderate-to-severe cGVHD was lower with PTCy (21.2% vs 27.3% p=0.461). Platelet engraftment was significantly delayed in the PTCy group (median 33 vs 21 days, p=0.030). Median time to neutrophil engraftment was also delayed in the PTCy group (20 vs 17 days, adjusted p=0.025). Age-adjusted HR for time to platelet engraftment was 0.60 (95% CI: 0.41–0.90; p=0.014), and for neutrophil engraftment 0.65 (95% CI: 0.44–0.95; p=0.025), indicating significantly delayed hematopoietic recovery with PTCy. Graft failure occurred in 9.4% vs 5.3% (p = 0.48); primary 7.5% vs 3.5%, secondary 1.9% vs 1.8% in PTCy vs non-PTCy groups, respectively. Relapse or progression occurred in 11.3% of PTCy vs 17.5% of others (p=0.355); four patients in the non-PTCy group progressed to AML post-transplant, versus none in the PTCy group (p=0.119). 1-year TRM was 22.6% in the PTCy group and 14.0% in the non-PTCy group, p = 0.324. Median OS was 72.8 months in the PTCy group (95% CI: 49.9 months, upper limit not estimable [NE]) and 93.4 months in the non-PTCy group (95% CI: 48.5 months, upper limit NE), although the difference was not statistically significant (p=0.739). After adjusting for age and HCT-CI, there were no significant differences in PFS (adjusted HR 1.37; 95% CI: 0.73–2.56; p=0.331) or OS (adjusted HR 1.27; 95% CI: 0.64–2.51; p=0.498). Conclusions In myelofibrosis patients undergoing allo-HCT, PTCy-based GVHD prophylaxis showed similar rates of grade ≥2 acute and moderate-severe chronic GVHD compared to non-PTCy regimens. Engraftment was significantly delayed, while OS and PFS were comparable. PTCy appears to be a safe and viable alternative to CNI-based prophylaxis.

e20629 Background: Lung cancer (LC) and respiratory failure (RF) are major contributors to mortality in adults globally. Understanding long-term mortality trends across demographic, racial, geographic, and socio-economic groups is crucial to addressing disparities and improving outcomes. This study examines national mortality patterns for LC and RF in the United States from 1999 to 2023, providing critical insights into changes over time and identifying key disparities in age-adjusted mortality rates (AAMRs). Methods: A descriptive analysis of mortality data from the CDC WONDER database (1999-2023) identified deaths due to LC (ICD-10 code C34) and RF (ICD-10 code J96) in adults aged 25+. AAMRs per 100,000 population were calculated and stratified by gender, race/ethnicity, region, state, and urban-rural classification. Temporal trends were analyzed with Joinpoint regression to compute APCs and 95% CIs. Statistical significance was set at p<0.05. The study followed STROBE guidelines. Results: From 1999 to 2023, a total of 456,325 adult deaths (age 25-85+) were attributed to Lung cancer and Renal failure. Age-adjusted mortality rates (AAMRs) for lung cancer (LC) and respiratory failure (RF) decreased from 1999 to 2010 (APC = -1.2, 95% CI: -1.6 to -0.9), followed by an increase from 2010 to 2023 (APC = 1.4, 95% CI: 1.1 to 1.7). Gender trends revealed a steady decline for men, while women's Age-Adjusted Mortality Rates (AAMr) initially decreased, then increased from 2011-2023 (APC: 2.0, 95% CI: 1.7 to 2.5). Non-Hispanic White populations experienced a rapid increase in age-adjusted mortality rates from 2014 to 2017 (APC: 2.7; 95% CI: 0.9 to 3.3), while other racial/ethnic groups exhibited varied trends with declines, increases, and stabilizations. Geographically, AAMRs were highest in the Northeast (8.6; 95% CI: 8.5 to 8.7) and lowest in the Midwest (7.1; 95% CI: 7.0 to 7.1). Metropolitan areas saw declining trends until 2006, stabilizing briefly, and increasing thereafter. Non-metropolitan areas experienced notable increases post-2009 (APC = 2.3, 95% CI: 1.8 to 3.1). State-level disparities revealed Mississippi with the highest AAMR (14.0; 95% CI: 13.7 to 14.4) and Wisconsin with the lowest (3.7; 95% CI: 3.5 to 3.8). These findings highlight significant demographic, racial, and geographic variations in LC and RF mortality trends. Conclusions: LC and RF-related mortality rates initially declined from 1999 to 2010 but increased significantly from 2010 onward, with notable demographic, racial, and geographic disparities. These findings highlight the urgent need for targeted interventions and resource allocation to address these rising trends and reduce inequities across vulnerable populations.