Browse Articles

Real-world incidence, prevalence, and short-term outcomes in young-onset breast cancer.

e13092 Background: Young-onset breast cancer(diagnosed at age < 40 years) is increasing in incidence,yet contemporary real-world (RW) estimates of age,race,and ethnicity stratified short-term survival remain incompletely characterized outside of population-based registries. Methods: A retrospective cohort study was conducted using the TriNetX Global Collaborative Network,an EHR platform.Female patients with incident BC(ICD-10-CM C50) diagnosed between 1/1/2015–1/1/2025 were included and stratified by age at diagnosis(18–39,40–49,50–64 years).Mortality incidence proportion was defined using all-cause mortality (“Deceased”) and further stratified by race and ethnicity.OS was evaluated over a fixed 3-year horizon to minimize bias from differential follow-up.Comparative analyses were performed using 1:1 propensity score matching (PSM) comparing age groups (18–39 vs 50–64; 40–49 vs 50–64; and 18–39 vs 40–49) with Kaplan–Meier estimation and Cox proportional hazards modeling.Sensitivity analyses included multivariable Cox proportional hazards models adjusting for comorbidities,stage,metastatic status and lines of treatment. Results: Incident cohorts included women 14,219 age18–39, 34,255 age 40–49,and 115,295 age 50–64.Mortality incidence proportion increased with age (18–39: 1.90%, 40–49: 2.16%, 50–64: 3.19%).Across all age groups, mortality incidence proportion was higher among Black vs White patients(18–39:3.40% vs 2.06%;40–49:3.70% vs 2.39%;50–64:5.16% vs 3.53%) and among non-Hispanic vs Hispanic patients (18–39:2.54% vs 1.73%;40–49: 2.45% vs 2.26%; 50–64: 3.74% vs 2.81%).In 3-year PSM analyses using ages 50–64 as the reference,all-cause mortality was lower for 18–39(HR 0.76, 95% CI 0.63–0.90;matched N = 12,327/cohort) and 40–49 (HR 0.88, 95% CI 0.78–0.98;matched N = 29,020/cohort),with no significant difference between 18–39 vs 40–49(HR 0.87, 95% CI0.72–1.04).Findings were consistent in adjusted Cox models (18–39 vs 50–64:aHR 0.78,95% CI0.68–0.89;40–49 vs 50–64:aHR 0.85, 95% CI0.78–0.93;18–39 vs 40–49:aHR 0.92, 95% CI0.79–1.07). Conclusions: In this large EHR-based cohort of incident BC,all-cause mortality increased with age and consistently differed by race and ethnicity,with higher mortality observed among Black and non-Hispanic patients across age strata.After PSM and multivariable adjustment, women diagnosed before age 50 experienced significantly lower short-term mortality risk compared with those aged 50–64.These RW findings provide updated context for age- and demographic-stratified outcomes and support targeted efforts to mitigate disparities through tailored surveillance and treatment strategies in younger BC populations. 3-year propensity score matched (PSM) overall survival. Comparison Matched N (per cohort) HR (95% CI) 18–40 vs 50–64 12,327 0.75 (0.63–0.90) 41–49 vs 50–64 29,020 0.87 (0.78–0.98) 18–40 vs 41–49 12,317 0.86 (0.71–1.04)

Breast cancer screening uptake and barriers among women in Hue, Vietnam: A cross-sectional survey.

e13530 Background: Breast cancer is the leading cancer in women in Vietnam, with almost 25,000 cases per year. However, there is limited data on breast cancer screening in Vietnam. We surveyed women in Hue, central Vietnam, to quantify screening uptake, knowledge, and modifiable barriers to breast cancer screening. Methods: We conducted an anonymous, cross-sectional survey of women in Hue, Vietnam, in 2025. The questionnaire assessed demographics, awareness of screening modalities, prior screening history (self-report), timing/location preferences, perceived benefits, barriers (personal and practical), and conditions that would increase willingness to screen. Descriptive statistics and multivariable regression analyses to identify factors associated with screening uptake were performed. Stepwise variable selection was implemented in R package MASS. Results: A total of 80 women participated in the survey, with the mean age of 53.3 years (range 40-71). Screening uptake was 46.2%. Only 13.8% (11/80) reported ever having mammography, while 37.5% (30/80) reported breast ultrasound. Knowledge gaps were prominent: 48.5% (38/80) reported never having heard of breast cancer screening methods; 25.0% (20/80) had heard of mammography. Most respondents believed screening helps detect cancer early (78.8%) and reduces mortality (67.5%), yet 36.2% felt screening is only needed when symptoms occur. The most common personal barrier was feeling healthy/no symptoms (55.0%), followed by fear of discovering cancer (8.8%) and never having heard about screening (8.8%). Practical barriers included high cost (15.0%), lack of time (10.0%), and lack of insurance coverage (5.0%). The most frequently endorsed facilitators were free/insurance-covered screening (82.5%), a physician reminder (52.5%), and availability at a nearby local health station (47.5%). In exploratory multivariable analysis, higher household income (≥11 million VND/month) and prior exposure to screening information were associated with higher odds of screening uptake. Conclusions: In this survey of women from Hue, Vietnam, fewer than half of women reported ever receiving breast cancer screening, and mammography uptake was low. Barriers were dominated by low perceived need in asymptomatic women, limited awareness, and cost/access constraints. Interventions combining education, insurance/financial support, and clinician-triggered reminders with more local availability may meaningfully increase screening uptake in central Vietnam.

Evaluation framework for monoclonal B-cell lymphocytosis.

e22571 Background: Monoclonal B cell lymphocytosis (MBL) is recognized as a precursor to chronic lymphocytic leukemia (CLL). There is an ongoing debate over whether MBL represents a benign, senescent condition or warrants further investigation and monitoring for potential progression. The natural history of MBL remains under scrutiny, and the need for surveillance in patients with incidental findings or familial risk factors is also being evaluated. Methods: We performed a systematic literature search using Web of Science and Science Direct databases to identify relevant studies. Our search using the term 'monoclonal B-cell lymphocytosis progression and monitoring' identified 844 documents. We selected articles if they were written in English and addressed monitoring or disease progression. Articles were excluded if published in other languages, were opinion pieces, or focused on treatment. Results: Our literature review identified 18 articles for qualitative analysis. These articles highlighted the importance of monitoring and assessing disease progression in MBL. Six articles examined the natural history of MBL, including its temporal behavior and progression to CLL requiring treatment. Four articles reported outcomes from patient cohorts monitored for 5 to 10 years. Five articles evaluated biomarkers, including genetic markers, imaging modalities, and cell counts, for their predictive value for disease progression and timing. Three articles provided specific monitoring recommendations, indicating that surveillance strategies should be based on whether MBL cell counts are classified as low or high. High MBL counts are associated with a 1-2% annual risk of progression, and annual clinical follow-up is recommended, including laboratory tests such as complete blood count (CBC), Fluorescence in situ hybridization (FISH), and assessment of markers such as immunoglobulin heavy-chain variable region (IGHV). The presence of IGHV was associated with a shorter time to first treatment. Emergence of new genetic mutations and deletions on chromosomes 17p or 11q were identified as potential indicators of progression. Physical examination remains essential for detecting B symptoms and increased infection frequency, which serve as indicators for further diagnostic evaluation. Conclusions: The literature review and qualitative analysis provide a comprehensive understanding of MBL indicating that high-risk MBL requires more extensive monitoring given progression risk. Progression occurs in approximately 1-2% of patients with high-count MBL. Longitudinal studies recommend monitoring these patients every 6 to 12 months, with baseline assessments including CBC, immunophenotyping, and a physical examination based on the 18 articles analyzed. Future research into timing and frequency of testing prognostic markers such as IGHV status and FISH is desired and would better define this condition and patients at higher risk of progression.

Discovery from single-cell RNA sequencing profiles of 18 CPTAC glioblastoma patients and validation in bulk profiles of 138 TCGA patients and two human-derived cell lines of a whole-transcriptome predictor of overall survival and drug targets by using quantum mechanics–based AI/ML.

3019 Background: Single-cell, more than bulk, multi-omic data, are small-cohort, noisy, and high-dimensional, and extremely difficult to model. We have developed artificial intelligence and machine learning (AI/ML) to overcome these challenges [doi: 10.1073/pnas.0530258100]. We have shown that our algorithms are uniquely able to discover accurate, precise, actionable, and mechanistically interpretable predictors, applicable to the general population, from the bulk multi-omes of as few as 19 patients [doi: 10.1158/1538-7445.AM2025-CT227]. We have demonstrated that these predictors of overall survival (OS) and drug targets consistently validate across laboratories and sometimes across indications, in federated and imbalanced studies and over time, and outperform all others where they exist [doi: 10.1063/1.5099268, 10.1200/JCO.2024.42.16_suppl.10043]. Here, we demonstrate our AI/ML in single-cell data. Methods: We used our algorithms to derive models from the single-cell RNA sequencing profiles of 18 glioblastoma (GBM) Clinical Proteomic Tumor Analysis Consortium (CPTAC) patients, and tested them in the bulk profiles of 138 patients in the Cancer Genome Atlas (TCGA). CPTAC and TCGA both profiled the core primary tumor of each patient, but CPTAC also sampled the peritumoral brain tissue. The two cohorts are indistinguishable in terms of their gender, age, and OS distributions, but they significantly differ in terms of race. Results: A whole-transcriptome model was discovered that is correlated with OS among the 18 CPTAC patients, with a Kaplan-Meier median OS difference of 30 months between the two groups of predictor-stratified patients, and a Cox hazard ratio of 4.6 and a concordance index of 87% (log-rank and Wald P-values < 5.0×10-2). When tested in the TCGA cohort, the model was a similarly significant predictor of OS. In both cohorts, despite the differences in profiling protocols and cohort demographics, the predictor outperformed the best standard-of-care indicator of OS in GBM, i.e., age. Consistent with a previous whole-genome predictor, which was derived from bulk DNA profiles, and validated in a clinical trial [doi: 10.1063/1.5142559], shorter OS was associated with overexpression of such anterior/posterior pattern specification genes as the Notch ligand DLL3 . We experimentally validated, in two human-derived cell lines, that its putative activator, METTL2A , is required for GBM cells' proliferation and viability [doi: 10.1158/1538-7445.AM2025-3686]. Conclusions: Our algorithms can discover predictors of patients’ OS and drug targets, in real-world small-cohort, noisy, and high-dimensional — single-cell — in addition to bulk multi-omic data, and the predictors experimentally validate.

Inside Front Cover: Enhancing Superlubricity and Wear Resistance in Mechanically Robust Hydrogel via Microliter‐Scale Subsurface‐Initiated Polymer Brush Grafting (Angew. Chem. Int. Ed. 23/2026)

Inter‐molecular Channel Gated Ion Migration for High Performance Zinc‐ion Batteries

ABSTRACT Aqueous zinc‐ion batteries (AZIBs) are plagued by dendrite growth and parasitic side reactions, which severely hinder their stable operation. Inspired by carrier proteins, amphiphilic surfactants are introduced as ion channel mediators. During zinc deposition, the cationic surfactants preferentially adsorb onto the zinc surface through their headgroups and self‐assemble into ordered ion channels, thereby guiding dense and uniform Zn 2+ flux. The ion channeling capability of cationic surfactants with various alkyl chain lengths is systematically investigated. Decyltrimethylammonium chloride, possessing the optimal chain length, is subsequently selected to construct efficient ion channels. Its moderate alkyl length enables the formation of stable ion channels that regulate the electric double layer (EDL) environment and promote directional zinc deposition. Benefiting from the ion‐regulating strategy, Zn||Cu asymmetric cells achieve an average Coulombic efficiency as high as 99.58% over 3000 cycles. Meanwhile, Zn||Zn symmetric cells operate stably for over 800 h under harsh conditions of 10 mA cm −2 with 10 mAh cm −2 . The assembled Zn||VO 2 pouch cell demonstrates impressive cycling stability over 300 cycles. This work provides a new perspective on surfactant‐enabled dendrite suppression and offers a low‐cost and highly effective solution for the practical development of AZIBs.

LaMnO3 perovskite for solar-driven degradation of coomassie brilliant blue dye

Eli Lilly inks dual-payload ADCs deal

STR-GNN: stability-regularized graph neural networks for suppressing spurious temporal variations in dynamic community detection

Abstract The study of temporal graphs frequently encounters spurious temporal fluctuations, wherein transient noise, inadequate observations, or ephemeral structural perturbations result in unstable and inconsistent community assignments. In dynamic networks, such variations may stem from measurement inaccuracies, sampling biases, or sudden yet non-informative changes in topology, leading graph neural networks (GNNs) to excessively respond to local fluctuations rather than accurately reflecting the genuine growth of communities. Consequently, the acquired representations may demonstrate considerable temporal inconsistency, resulting in community structures that fluctuate erratically across successive time intervals. This instability significantly constrains the reliability of dynamic community recognition in temporal GNNs, especially in streaming, partially observed, or weakly supervised contexts, where the model must perpetually adjust to changing graph structures without comprehensive ground-truth supervision. In these circumstances, the learning process may provide community assignments that seem credible at each moment yet lack overall temporal consistency. We refer to this issue as spurious temporal variation, where temporal GNNs produce unstable community assignments caused by transient perturbations rather than genuine structural evolution. The suggested method mitigates false temporal fluctuations by implementing stability-aware regularization and consistency restrictions across successive graph snapshots, promoting the model’s ability to maintain coherent community structures while adapting to authentic structural changes. The proposed framework enhances the resilience, reliability, and interpretability of temporal GNN-based community identification by stabilizing community evolution over time, hence rendering it more appropriate for real-world dynamic networks characterized by noise, perturbations, and incomplete data.

Histone H2B-associated proteins: The Arabidopsis nucleolin 1 binds H2B and facilitates nucleosome disassembly via RNA-dependent mechanism

Clinical characteristics and neighborhood factors associated with patient survival among adolescents and young adults with bone sarcoma.

e23500 Background: Compared to pediatric patients, adolescents and young adults (AYA) with bone sarcomas experience clinical and socioeconomic challenges that leads to poorer survival outcomes. Socioeconomic challenges can be measured by Area Deprivation Index (ADI) which is a validated census block- group tool that ranks neighborhoods. We sought to evaluate the association between ADI and clinical outcomes for AYA patients with bone sarcomas treated at a tertiary cancer center. Methods: A retrospective cohort was curated of 292 AYA patients (15 to 39 years old) with bone sarcoma treated between 1997 to 2023. To identify factors associated with patient outcomes, survival analyses were conducted by demographic characteristics, histology, treatment, and neighborhood ADI. ADI was derived from patients’ residential addresses at diagnosis and linked to neighborhood-level deprivation scores. ADI was analyzed as a categorical variable grouped into quartiles, with higher quartiles representing greater neighborhood disadvantage. Results: Among this cohort of AYA patients, the most common diagnosis being osteosarcoma (62.0%). Survival rates for all patients at 12, 36, and 60 months were 90.0%, 70.1%, and 60.3%, respectively. Local recurrence (LR) occurred among 24.3% of the patients, and the median time to recurrence was 19.4 months (IQR 11.9 - 48.3). Overall survival and LR did not significantly differ by age, sex assigned at birth, race, ethnicity, marital status, insurance status, tumor site, alcohol use, education level, or employment status. ADI at the federal level was associated with 60-month survival, with patients residing in higher deprivation areas demonstrating worse survival compared with those in lower deprivation areas (53.7% vs. 66.3%, p =0.04). Conclusions: Among AYA osteosarcoma patients, long-term survival is driven primarily by tumor biology and initial stage, rather than age, sex, race, lifestyle factors. Patients living in more socioeconomically deprived neighborhoods (higher federal ADI) experienced significantly worse overall survival. Advanced stages and the need for complex, multimodal therapy mark the steepest survival declines. These findings underscore the need for strategies focused on access to multidisciplinary care and targeted support for patients residing in socioeconomically disadvantaged communities.

Beyond imatinib: Real-world mortality patterns across modern second-line TKIs in CML.

e18594 Background: Tyrosine kinase inhibitors (TKIs) have altered management and outcomes of chronic myeloid leukemia (CML). Imatinib has been a common first-line TKI for several decades, but some patients discontinue this agent due to intolerance or poor hematologic or molecular response and are switched to second-line TKIs including nilotinib, bosutinib, dasatinib, or ponatinib depending on side effect profiles and mutational analysis. Randomized trials have established the effectiveness of second-line TKIs, but comparative real-world survival outcomes after imatinib discontinuation are not as established. Methods: We conducted a retrospective cohort study using the TriNetX Global Collaborative Network, which includes electronic health records from several healthcare organizations. Adult patients (>18 years) with BCR-ABL positive CML who received imatinib and subsequently started on a second-line TKI were identified. The index date was defined as initiation of the second-line agent after imatinib exposure. Patients treated with dasatinib, bosutinib, ponatinib, or asciminib were compared with those receiving nilotinib. Nilotinib was selected as the reference as it had the largest available cohort size. The primary outcome was all-cause mortality. Pairwise comparisons were performed using 1:1 propensity score matching for demographics and major comorbidities. Survival was assessed using Kaplan-Meier analysis with hazard ratios and log-rank testing. Results: After matching, cohort sizes ranged from 320 to 662 patients in each group. Mortality did not significantly differ between dasatinib and nilotinib (15.5% vs 14.8%; HR 1.17) or between bosutinib and nilotinib (21.5% vs 23.6%; HR 1.07). Asciminib cohort was associated with lower mortality (9.2% vs 21.8%), but Kaplan-Meier survival analysis did not reveal a statistically significant difference. In contrast, ponatinib was associated with higher mortality compared with nilotinib (28.8% vs 20.0%; HR 2.00, 95% CI 1.45–2.76; p<0.001). After matching, baseline characteristics were well balanced. Conclusions: In this large real-world cohort of patients with CML who discontinued imatinib, survival outcomes after second-line TKI therapy were similar among dasatinib, bosutinib, and nilotinib, while asciminib illustrated lower mortality. The higher mortality seen with ponatinib could be due to the differences in underlying mutation profiles or this drug's side effects. These findings provide real-world data which supports existing trials and reassures individual selection of second-line TKIs after discontinuing imatinib.

The AI paradox in precision oncology: Prospective blinded validation of large language models against molecular tumor board.

11047 Background: Molecular tumor boards (MTBs) are central to precision oncology but remain limited in accessibility. Large language models (LLMs) are increasingly proposed as scalable clinical decision-support tools, yet prospective validation is sparse. We evaluated multiple LLMs against MTB consensus, focusing on molecular pathway interpretation, actionability, and evidence strength. Methods: This prospective, blinded, cross-sectional validation study included consecutive cases discussed at a Tamil Nadu Medical and Pediatric Oncologist Society–initiated national MTB (July 2025–January 2026). Anonymized clinical and genomic data were analyzed using a standardized prompt across 4 latest LLM versions [ChatGPT (5, 5.1, 5.2), Perplexity, Gemini (2.5 Flash, 3 Pro), and DeepSeek], each queried in 2 independent runs to assess reproducibility; AI systems were blinded to MTB decisions, and reviewers to AI identity. Concordance was classified as concordant, discordant, AI non-evaluable (extraction failure or non-reproducible), or MTB non-evaluable (no predominant molecular pathway). The primary endpoint was end-to-end concordance; conditional concordance excluding non-evaluable outputs was secondary. Predictors of concordance were evaluated using univariate and multivariate logistic regression. Results: Of 108 cases, 80 were evaluable for AI comparison. Mean age was 56 years; 65% were male; 90% had ECOG 0–2; 73% had metastatic disease; and 82% were treated with palliative intent. The cohort was heavily pretreated, with 48% receiving ≥3 prior lines. Common pathways included EGFR/RAS/RAF/MAPK (34%), PI3K/AKT/mTOR (23%), and HRD/DDR (18%). End-to-end concordance was 88%, 74%, 83%, and 55% across the four LLMs. AI non-evaluable outputs due to extraction failure or non-reproducibility across two independent runs occurred in 3.8% to 21.3% of cases across platforms, highlighting important limitations in reliability. Citation-level hallucination rates were high (41%, 36%, 49%, and 48%). On univariate analysis, higher evidence level predicted concordance across all models (all P ≤ 0.03). ESCAT tier was also significantly associated with concordance for LLM-2, LLM-3, and LLM-4. On multivariate analysis, evidence level remained the only consistent independent predictor for three LLMs, while ESCAT tier retained significance for one model. Inter-rater reliability was excellent (96.3%; κ = 0.93). Conclusions: LLM concordance is highest in guideline-supported, high-actionability settings but declines in low-evidence, poor-targetability scenarios—precisely where clinical support is most needed. This “AI paradox,” combined with substantial hallucination risk, indicates that while LLMs may assist first-pass molecular interpretation, expert multidisciplinary MTBs remain essential for safe and reliable precision oncology.

Effect of GLP-1 receptor agonism on anthracycline- and HER2-targeted therapy–induced cardiac dysfunction and systemic inflammation in non-diabetic preclinical models.

e12524 Background: Sequential administration of anthracyclines and HER2-targeted agents represents a highly effective therapeutic strategy in breast cancer but is frequently complicated by cumulative cardiotoxicity, limiting treatment durability and long-term cardiovascular outcomes. Inflammation-driven myocardial injury and multi-organ stress responses are increasingly recognised as central mechanisms in cancer therapy–related cardiac dysfunction. Glucagon-like peptide-1 receptor agonists (GLP-1RAs) exert pleiotropic cardiovascular and anti-inflammatory effects independent of glucose lowering, yet their role in preventing chemotherapy-induced cardiac injury in non-diabetic settings remains largely unexplored. Methods: Non-diabetic murine models were exposed to sequential doxorubicin followed by trastuzumab over a 10-day protocol. Animals were randomised to receive the GLP-1 receptor agonist semaglutide or vehicle. Cardiac function was serially assessed by echocardiography, including left ventricular ejection fraction and myocardial strain parameters. Histopathological analyses evaluated myocardial fibrosis and cardiomyocyte hypertrophy. Immunohistochemical profiling was performed in cardiac, hepatic, and renal tissues. Circulating biomarkers of myocardial injury and systemic inflammation, including troponins, IL-1β, IL-6, IL-8, CCL12, and high-sensitivity C-reactive protein, were quantified. Results: Sequential anthracycline and HER2 blockade induced marked cardiac dysfunction, characterised by significant impairment of systolic performance and myocardial deformation, accompanied by increased myocardial fibrosis, cardiomyocyte hypertrophy, and elevated circulating markers of cardiac injury and inflammation. Semaglutide administration significantly preserved left ventricular ejection fraction and strain parameters and substantially attenuated structural myocardial remodelling. These functional benefits were paralleled by a robust reduction in circulating troponins and pro-inflammatory cytokines. In addition, semaglutide mitigated inflammatory signalling across cardiac, hepatic, and renal tissues, indicating a broader cardio-renal protective profile. Conclusions: These findings support a novel, glucose-independent role for GLP-1RAs in cardio-oncology, with potential implications for integrated strategies aimed at preserving cardiovascular and multi-organ health during intensive anticancer treatment. Further translational studies are warranted to define their clinical relevance in patients at high risk of cancer therapy–related cardiac dysfunction.

First-line NALIRIFOX in patients with metastatic pancreatic ductal adenocarcinoma and pre-existing diabetes: NAPOLI 3 post hoc analysis.

4195 Background: Approximately 29% of patients with metastatic pancreatic ductal adenocarcinoma (mPDAC) present with pre-existing diabetes. Diabetes increases the risk of peripheral neuropathy (PN), raising concerns that oxaliplatin-containing regimens may exacerbate PN risk. NALIRIFOX (liposomal irinotecan, oxaliplatin, leucovorin and 5-fluorouracil) is approved for first-line (1L) treatment of mPDAC based on significantly improved overall survival versus nab-paclitaxel plus gemcitabine (Gem+NabP) in the phase 3 NAPOLI 3 trial (hazard ratio 0.83; 95% confidence interval [CI] 0.70–0.99). The overall response rate (ORR) for NALIRIFOX was 41.8% (95% CI: 36.8–46.9%) and no unexpected safety concerns were identified. This post hoc exploratory analysis of NAPOLI 3 evaluates outcomes including PN development among patients with pre-existing diabetes. Methods: Patients randomized to 1L NALIRIFOX in NAPOLI 3 who had pre-existing diabetes were assessed, with subgroups according to whether or not they developed PN. The primary endpoint was ORR; patient characteristics and adverse event rates were also described. No hypotheses were tested owing to small sample sizes; results are descriptive. Results: Of 383 patients randomized to receive NALIRIFOX in NAPOLI 3, 96 (25%) had pre-existing diabetes, of whom 94 were included in the safety population. PN developed in 13/94 (14%) patients (Grade ≥3: 2/13 [15%]) with pre-existing diabetes, none of whom had evidence of PN at baseline. Among patients without diabetes, PN developed in 53/276 (19%; Grade ≥3: 10/53 [19%]). Patients with pre-existing diabetes who developed PN had a lower median age, lower proportion of males, higher metastatic burden, and higher frequency of tumors in the body of the pancreas than patients who did not develop PN (Table). Among patients with pre-existing diabetes, ORR for NALIRIFOX was 43.8% (42/96 [95% CI: 33.6–54.3%]); all responses were partial. ORR was 69.2% (9/13 [95% CI: 38.6–90.9%]) among patients with pre-existing diabetes who developed treatment-emergent PN and 40.7% (33/81[95% CI: 29.9–52.2%]) among those who did not (Table). Conclusions: In this exploratory analysis, PN incidence and clinical outcomes in patients with pre-existing diabetes were consistent with those without diabetes and with the overall NALIRIFOX population. In this small population, pre-existing diabetes did not appear to put patients at risk of developing treatment-emergent PN or to impact treatment outcomes. Clinical trial information: NCT04083235 . Patients with baseline diabetes who developed peripheral neuropathy(n = 13) Patients with baseline diabetes who did not develop peripheral neuropathy(n = 81) Baseline characteristics Age, median, years 63.0 66.0 Male, n (%) 6 (46.2) 52 (64.2) ≥ 3 metastatic sites, % 53.8 40.7 Tumor in body of pancreas, % 38.5 22.2 Efficacy ORR, % (95% CI) 69.2 (38.6–90.9) 40.7 (29.9–52.2)

Metabolic comorbidity burden and oncologic features of early-onset colorectal cancer stratified by hepatic steatosis.

e15696 Background: Early-onset colorectal cancer (EO-CRC) incidence is rising, and emerging evidence links EO-CRC with metabolic dysfunction and hepatic steatosis. As CRC staging routinely includes abdominal imaging, this setting offers an opportunity to quantify pre-treatment hepatic steatosis. We report the prevalence of metabolic comorbidities in EO-CRC and compare clinicopathologic features stratified by hepatic steatosis status in a single institution cohort. Methods: We performed a retrospective review of patients aged 18-49 diagnosed with CRC (2019-2025). Records were abstracted for demographics, comorbidities, and tumor characteristics. Hepatic steatosis was assessed using radiology and pathology reports and compared to ICD-9/10 codes. Group comparisons utilized nonparametric tests for continuous variables and chi-square or Fisher’s exact tests for categorical variables. Results: A total of 613 patients were identified (median age 44.0 years [IQR 40.0–48.0]; 50.7% female). Metabolic comorbidities were common: obesity in 38.5%, hypertension in 37.7%, hyperlipidemia in 26.4%, and type 2 diabetes in 16.5%. Adenocarcinoma was the predominant histology (96.8%). Among patients with available staging, advanced disease was frequent with 70.4% presenting with stage III–IV disease, 71.5% were left-sided, and 7.6% exhibited deficient mismatch repair. In the tumor registry subset with comprehensive chart review (n = 258 after exclusions), hepatic steatosis was identified in 45.0% (116/258). Notably, 75.9% (88/116) of patients with steatosis lacked a corresponding ICD diagnosis. Steatosis was present at or within three months of cancer diagnosis in 51.7% (60/116) of cases. Compared to patients without steatosis at diagnosis (n = 198), those with steatosis had significantly higher median BMI (33.9 vs 27.3 kg/m²; p < 0.001), obesity (71.7% vs 38.9%; p < 0.001), hypertension (53.3% vs 31.3%; p = 0.003), and obstructive sleep apnea (26.7% vs 8.6%; p < 0.001). Stage at diagnosis differed between groups (p = 0.047), with a lower proportion of stage IV disease observed in the steatosis group (18.3% vs 32.8%; unknown stage 16.7% vs 6.6%). Tumor characteristics (histology, sidedness, and MMR) were similar between groups. Conclusions: Hepatic steatosis is highly prevalent in EO-CRC but remains under-recognized in clinical coding. While patients with steatosis exhibit higher metabolic burden, tumor characteristics are similar. The lower proportion of metastatic disease in the steatosis cohort warrants further investigation. These findings highlight the need for systematic identification of Metabolic Dysfunction-Associated Steatotic Liver Disease (MASLD) in EO-CRC for risk stratification and survivorship. Additional analyses, including mutational profiling, sites of metastasis and preliminary survival outcomes are underway and will be presented.

Comparative outcomes of [^177Lu]Lu-PSMA-617 versus radium-223 in bone-metastatic castration-resistant prostate cancer: A real-world propensity-matched analysis.

e17070 Background: Two guideline-recognized radiopharmaceuticals differ in mechanism and indication. Radium-223 (Ra-223), an alpha-emitter targeting osteoblastic lesions, and Lu-PSMA-617, a PSMA-targeted beta-emitter active in bone and soft-tissue metastases; both improved survival in their respective pivotal trials. Lu-PSMA-617 received FDA approval for taxane-naïve mCRPC after ARPI progression (PSMAfore). Despite these advances, no head-to-head data exist. Clinicians often choose between Ra-223 and Lu-PSMA-617 for mCRPC when chemotherapy is unsuitable. Methods: We conducted a retrospective study using the TriNetX Network of adult men with prostate cancer and bone metastases who received Lu-PSMA-617 or Ra-223. Castration resistance was approximated by ongoing ADT and prior ARPI. Patients with visceral metastases were excluded. Outcomes were assessed at 3, 6 & 12 months. Survival was analyzed with Kaplan–Meier methods, log-rank tests, and Cox models to derive hazard ratios (HR) with 95% CI. Results: At 3 months, mortality was substantially lower with Lu-PSMA-617 (HR 0.34, 95% CI 0.18–0.6, p = 0.001). Hospitalization was also reduced (HR 0.46, p = 0.035). Ra-223 patients had higher rates of severely elevated PSA levels (≥50 ng/mL), yielding (HR 0.57, p = 0.07).At 6 months, the survival benefit persisted (HR 0.37, 95% CI 0.25–0.56, p < 0.001). Hospitalizations were again fewer with Lu-PSMA-617 (HR 0.47, p = 0.009). ER visits did not differ significantly (HR 0.65, p = 0.199). PSA outcomes favored Lu-PSMA-617 (HR 0.57, p = 0.025). Hematologic safety endpoints were neutral.At 1 year, Lu-PSMA-617 demonstrated a marked survival benefit (HR 0.46, 95% CI 0.35–0.61, p < 0.001). Hospitalizations were fewer with Lu-PSMA-617 (HR 0.56, p = 0.01). ER visits were similar between groups (HR 1.04, p = 0.71). Rates of PSA ≥50 ng/mL were less frequent in the Lu-PSMA-617 group (HR 0.56, p = 0.006). Hematologic outcomes remained neutral. Conclusions: In this real-world propensity-matched analysis, Lu-PSMA-617 was associated with superior survival and fewer hospitalizations compared with Ra-223 in bone-metastatic mCRPC. While Ra-223 remains a preferred option per NCCN for symptomatic bone-predominant disease without visceral metastases, our findings suggest that when PSMA-avid disease is present, Lu-PSMA-617 should be prioritized as the radiopharmaceutical of choice. Prospective studies are warranted to confirm optimal sequencing.

Pancreatectomy with and without preoperative biliary drainage: A systematic review and meta analysis.

e16426 Background: Preoperative biliary drainage (PBD) is commonly performed in patients with obstructive jaundice due to pancreatic or biliary tumors prior to pancreatectomy. While intended to reduce perioperative complications, its clinical benefit remains controversial, with concerns regarding increased infection and morbidity. This systematic review and meta-analysis aimed to evaluate the impact of PBD on postoperative outcomes, mortality, and perioperative parameters in patients undergoing pancreatectomy. Methods: We conducted a systematic search of PubMed, Embase, and Cochrane Library for randomized controlled trials, non-randomized trials, and observational studies comparing pancreatectomy with PBD versus pancreatectomy alone. Nine studies encompassing 6,572 patients (4,291 in the PBD group; 2,281 in the control group) met inclusion criteria. Outcomes analyzed included overall and major complications, surgery-related complications (postoperative pancreatic fistula [POPF], biliary leakage, surgical site infection, pneumonia, thromboembolic events), bile culture positivity, reoperation rates, 30-day and in-hospital mortality, operative time, blood loss, transfusion requirements, and total hospital stay. Pooled odds ratios (OR) or mean differences (MD) with 95% confidence intervals (CI) were calculated using random-effects models. Results: PBD was associated with a higher risk of major complications (OR 1.22; 95% CI 1.03–1.44) and significantly increased bile culture positivity (OR 14.02; 95% CI 9.74–20.19). Overall complications (OR 1.10; 95% CI 0.98–1.25), POPF (OR 1.08; 95% CI 0.89–1.31), biliary leakage (OR 0.67; 95% CI 0.47–1.13), surgical site infection (OR 1.08; 95% CI 0.89–1.29), pneumonia (OR 1.11; 95% CI 0.82–1.50), thromboembolic events (OR 0.86; 95% CI 0.60–1.23), and reoperation rates (OR 1.00; 95% CI 0.76–1.31) were not significantly different. In-hospital mortality was higher in the PBD group (OR 3.32; 95% CI 1.28–8.66), whereas 30-day mortality did not differ significantly (OR 1.31; 95% CI 0.82–2.07). PBD increased operative time (MD 12.04 min; 95% CI 9.27–15.54), intraoperative blood loss (MD 78.94 mL; 95% CI 65.67–92.20), and total hospital stay (MD 4.44 days; 95% CI 4.08–4.81), without significantly affecting transfusion requirements (MD –0.35 units; 95% CI –0.81–0.10). Conclusions: Preoperative biliary drainage prior to pancreatectomy is associated with increased major complications, higher bile contamination, longer operative time, and prolonged hospital stay, without significant improvement in 30-day mortality or overall postoperative complications. These findings suggest careful patient selection for PBD, reserving it for cases with severe jaundice or cholangitis.

Where do adults with pancreatic cancer die in the United States: A retrospective analysis of 25 years.

e24042 Background: Pancreatic cancer remains a leading cause of cancer-related morbidity and mortality. With limited survival for many patients, end-of-life care is a critical component of pancreatic cancer management. Despite the emphasis on palliative care, information regarding where adults with pancreatic cancer spend their final days remains largely unexplored. This study aimed to analyze the location and circumstances of death among adults with pancreatic cancer in the United States using publicly available data from the Centers for Disease Control and Prevention's Wide-ranging Online Data for Epidemiological Research (CDC WONDER). Methods: Mortality data of individuals with pancreatic cancer aged > 25 years was obtained from CDC WONDER and analyzed from 1999 to 2023. The places of death were categorized as the medical facility, descendants’ home, hospice facility, nursing homes/long-term care facility. Results: A total of 956,029 deaths were reported in individuals with pancreatic cancer aged > 25 years from 1999 to 2023. Pancreatic cancer-related age-adjusted mortality rates per 100,000 increased from 1999 (16.41) to 2023 (17.17). Descendants' homes experience the greatest number of deaths (49.21%, 470,448) from 1999 to 2023. Deaths at home nearly doubled, rising from 1.37% in 1999 (13,139) to 2.77% in 2023 (26,464). Medical facility deaths increased from 1.04% in 1999 (9,903) to 1.12% in 2023 (10,731). Deaths at hospice facilities increased from 0.02% in 2003 (214) to 0.66% in 2023 (6,267). Similarly, deaths at other locations increased from 0.19% (1,781) in 1999 to 0.23% (2,204) in 2023 Deaths in nursing homes/long-term care declined from 0.44% in 1999 (4245) to 0.39% in 2023 (3,769). Unknown locations accounted for a total of 2050 deaths. Conclusions: This study highlights the changing patterns in the places of death among patients with pancreatic cancer, with the most deaths occurring at descendants' homes and a twofold increase in pancreatic cancer-related deaths occurring at home over the past two decades. Further research is needed to determine the end-of-life needs and preferences of these patients and their care-takers and identify the basis of these disparities. Place of death Years Deaths (n) % of total Decedent’s home 1999–2023 470,448 49.21 Medical facility 1999–2023 232,458 24.31 Nursing home/LTC 1999–2023 107,265 11.22 Hospice facility 2003–2023 90,745 9.49 Other/Unknown 1999–2023 55,113 5.77 Total 1999–2023 956,029 100

Impact of hypofractionated radiotherapy on pathologic response in locally advanced cervical cancer.

e17518 Background: Cervical cancer remains a significant global health burden, particularly in low- and middle-income countries, where limited radiotherapy infrastructure restricts access to standard treatment. Although concurrent chemoradiotherapy is the standard of care, hypofractionated radiotherapy may improve treatment efficiency while maintaining oncologic outcomes. Our purpose is to compare pathologic response outcomes between hypofractionated and standard radiotherapy in patients with locally advanced cervical cancer. Methods: Pathologic response was evaluated in patients with FIGO stage IB3–IIIC1 cervical cancer enrolled in a phase II randomized trial. Patients received concurrent chemotherapy with either standard external-beam radiotherapy (50 Gy in 25 fractions) or hypofractionated radiotherapy (37.5 Gy in 15 fractions), followed by radical hysterectomy and bilateral pelvic lymphadenectomy. Pathologic response and lymph node involvement were compared between groups using chi-square or Student’s t tests, as appropriate. A p value < 0.05 was considered statistically significant. Results: Ninety-three patients were included (50 standard fractionation; 43 hypofractionation). Baseline clinical and tumor characteristics were balanced between groups. Squamous cell carcinoma was the most common histology (84%). Complete pathologic response was observed in 48% of patients in the standard arm and 35% in the hypofractionated arm, while residual microscopic disease was identified in 38% and 44%, respectively ( p = 0.46). Additional treatment due to partial response was required in 14% of patients receiving standard fractionation and 21% receiving hypofractionation. Adverse pathologic features prompting further treatment included residual tumor >1 cm, lymphovascular space invasion, deep stromal invasion, and positive surgical margins. The mean number of lymph nodes resected was similar between groups (14 vs. 15; p = 0.291). Among 1,399 lymph nodes examined, 0.85% were positive on final pathology. Conclusions: Pathologic response following hypofractionated radiotherapy appeared comparable to that achieved with standard fractionation. While hypofractionation is not currently standard for cervical cancer, these findings suggest it may represent a feasible alternative in settings with limited radiotherapy capacity. Larger studies with longer follow-up are required to further assess efficacy, safety, and pathological response patterns. Clinical trial information: NCT03750539 .