Browse Articles

Efficacy and toxicity of CDK4/6 inhibitors in early and metastatic HR+/HER2– breast cancer.

e13047 Background: Cyclin-dependent kinase 4/6 (CDK4/6) inhibitors improve outcomes in metastatic hormone receptor-positive, human epidermal growth factor receptor 2– negative (HR+/HER2–) breast cancer (MBC), but their benefit in early-stage disease and the efficacy-toxicity balance remain uncertain. We conducted a meta-analysis of phase III trials comparing CDK4/6 inhibitors plus endocrine therapy (ET) versus ET alone. Methods: MEDLINE, Embase, and Web of Science were searched through October 1, 2025. Phase III randomized trials comparing CDK4/6 inhibitors plus ET versus ET alone were included. Two reviewers independently extracted data and assessed risk of bias using the Cochrane Risk of Bias 2.0 tool. Random-effects models with Hartung–Knapp adjustment pooled hazard ratios (HRs) for time-to-event outcomes and risk ratios (RRs) for dichotomous outcomes. Statistical heterogeneity was assessed using the I 2 statistic, with values > 50% reported in the results. Primary outcomes were overall survival (OS) and progression-free survival (PFS); secondary outcomes included invasive disease-free survival (iDFS), distant relapse-free survival (DRFS), objective response rate (ORR), clinical benefit rate (CBR), disease control rate (DCR), treatment discontinuation (TDR), dose reduction (DRR), and grade 3–4 adverse events. The study followed PRISMA 2020 guidelines (PROSPERO: CRD420251132302). Results: Twenty-two trials were included (18 MBC, n = 6,364; 4 EBC, n = 17,741; outcomes reported variably). In MBC, CDK4/6 inhibitors improved OS (HR 0.78, 95% CI 0.72–0.85; p < 0.001) and PFS (HR 0.53, 95% CI 0.50–0.56; p < 0.001), with consistent OS benefit in ET-naive (HR 0.80, 95% CI 0.71–0.89; p < 0.001) and previously ET-treated patients (HR 0.75, 95% CI 0.61–0.93; p = 0.009). ORR (RR 2.07, 95% CI 1.26–3.40; p = 0.004; I 2 = 92.5%), CBR (RR 1.25, 95% CI 1.18–1.32; p < 0.001; I 2 = 72.9%), and DCR (RR 1.09, 95% CI 1.04–1.15; p = 0.001; I 2 = 57.1%) were increased. TDR and DRR were increased (RR 2.39, 95% CI 1.54–3.70; and RR 9.51, 95% CI 6.96–13.01; both p < 0.001). Hematologic and non-hematologic adverse events (diarrhea, liver enzymes, infections) were increased. In EBC, iDFS improved (HR 0.81, 95% CI 0.67–0.98; p = 0.029; I 2 = 60%), while OS (HR 0.92, 95% CI 0.75–1.12; p = 0.40) and DRFS (HR 0.81, 95% CI 0.24–2.67; p = 0.73; I 2 = 73.2%) were not significant. TDR and DRR were increased (RR 8.39, 95% CI 2.24–31.42; p = 0.014; I 2 = 95.7%; and RR 26.84, 95% CI 14.85–48.49; p < 0.0001). Hematologic and non-hematologic adverse events (fatigue, liver enzymes) were increased. Conclusions: CDK4/6 inhibitors confer significant OS and PFS benefits in MBC, supporting standard first-line use. In EBC, iDFS improved without demonstrated OS benefit and was accompanied by increased treatment modifications and toxicity, underscoring the need for individualized risk-benefit assessment. Longer follow-up is warranted to clarify immature OS data.

Effects of a multimodal TCR/BCR repertoire foundation model on blood RNA-seq–based prediction of severe adverse event risk and rheumatoid arthritis.

2570 Background: Severe immune-related adverse events (irAEs) limit the use of immune checkpoint inhibitors (ICIs). Clinically, irAEs can resemble autoimmune diseases such as rheumatoid arthritis (RA), suggesting shared dysregulation of adaptive immune receptor repertoires (T-cell receptor [TCR] and B-cell receptor [BCR]). We developed a TCR/BCR foundation model to predict severe irAE risk prior to ICI treatment and to detect RA from peripheral blood. Methods: Peripheral blood RNA sequencing (RNA-seq) from cancer patients prior to ICI therapy, healthy donors, and RA patients was analyzed across six cohorts (Table 1). Severe irAEs were defined as grade ≥3. Gene expression quantification and signature scores were computed using kallisto and ssGSEA. Adaptive immune receptor repertoire (AIRR) features for TCR and BCR were derived from blood RNA-seq using an in-house pipeline (pyigmap). We trained neural network models to (i) estimate severe irAE risk in ICI-treated cancer patients and (ii) distinguish RA from healthy donors, using TCR/BCR embedding features alone or combined with gene signature scores. Performance was evaluated across multiple datasets to assess discrimination and cross-cohort robustness. Results: The model combining TCR/BCR embedding features with gene signature scores achieved improved discrimination for severe irAE risk prediction (AUC 0.78; p=1×10⁻⁵) and RA detection (AUC 0.70; p=5×10⁻¹⁴). This outperformed models based on gene signatures alone (irAE AUC 0.67; p=8×10⁻³; RA AUC 0.66; p=4×10⁻⁹), TCR/BCR diversity alone (irAE AUC 0.66; p=2×10⁻²; RA AUC 0.61; p=6×10⁻⁵), or signatures plus diversity (irAE AUC 0.70; p=2×10⁻³; RA AUC 0.69; p=9×10⁻¹³). Embedding-based representations were more robust to technical variation than conventional diversity metrics: batch-effect decomposition showed higher biological variability (0.012 vs 0.002) and lower technical variability (0.460 vs 0.861) for embeddings versus diversity features, consistent with improved cross-cohort generalization. Conclusions: A TCR/BCR foundation model integrated with gene signatures improves pre-treatment prediction of severe irAE risk and detection of autoimmune disease from peripheral blood RNA-seq. These repertoire-informed representations capture clinically relevant clonotype biology and support minimally invasive risk stratification and monitoring for patients receiving ICIs. Cohort description. Cohort Source N Severe irAE No severe irAE #1 (irAE Train Set 1) Internal 726 48 678 #1 (irAE Test Set 1) Internal 187 12 175 #2 (irAE Test Set 2) Internal 47 10 37 Cohort Source N Healthy RA #3 (RA Train Set 1 / Validation Set) Open source 211 97 114 #4 (RA Test Set 1) Open source 101 50 51 #5 (RA Test Set 2) Open source 24 12 12 #6 (RA Test Set 3) Open source 140 20 120

Cardiac substructure dose signatures as predictors of late cardiovascular events in childhood cancer survivors.

12123 Background: Radiation-associated cardiovascular disease (CVD) remains a major late toxicity in childhood cancer survivors. Whole-heart dose may mask heterogeneous substructure exposure, limiting precision in survivorship surveillance and modern radiotherapy planning. Methods: Using the Childhood Cancer Survivor Study (CCSS) open-access cardiac substructure dosimetry dataset (12 mean-dose structures; Gy), we modeled time from cancer diagnosis to CAD, CHF, valvular disease, and arrhythmia with elastic-net Cox (α=0.5; 10-fold CV). Bootstrap stability selection (200 resamples) quantified reproducible substructure predictors; “core” structures were defined by selection frequency ≥0.80. Parsimonious Cox models were fit using core structures (dose scaled per 5 Gy) and adjusted for sex, race, age at diagnosis, anthropometric category, and smoking. Model discrimination was assessed using the concordance index (C-index). Model-based predicted 20-year absolute risks were used to estimate dose thresholds for clinically interpretable risk targets (reference profile: median covariates; modal factor levels). Results: Among 25,481 survivors, events occurred for CAD (n=475), CHF (n=544), valvular disease (n=131), and arrhythmia (n=181). LAD and aorta emerged as reproducible, endpoint-specific predictors: LAD was core across CAD/CHF/valvular outcomes; aorta and LMCA were core for CAD/valvular; pericardium was core for CAD/CHF. In adjusted core models, aortic dose was associated with CAD (HR 1.59 per 5 Gy; 95% CI 1.14–2.21; C-index 0.796) and valvular disease (HR 1.96; 95% CI 1.15–3.32; C-index 0.773). LAD dose was associated with CAD (HR 1.26; 95% CI 1.05–1.53; C-index 0.796) and CHF (HR 1.23; 95% CI 1.02–1.47; C-index 0.818). Absolute risk translation identified interpretable thresholds: predicted 20-year CAD risk crossed ~1% at 13 Gy mean aortic dose and ~2% at 20 Gy; CAD risk crossed ~1% at 24 Gy mean LAD dose. Predicted 20-year CHF risk crossed ~0.5% at 21 Gy mean LAD dose. Predicted 20-year valvular risk crossed ~1% at 23 Gy mean aortic dose. Conclusions: Substructure-specific dosimetry yields reproducible radiation “signatures,” with LAD and aorta dose consistently linked to late CVD. These dose–risk thresholds provide actionable benchmarks for risk-stratified survivorship surveillance and may inform substructure-sparing dose constraints in contemporary radiotherapy planning moving beyond whole-heart dose to endpoint-specific substructure constraints. Substructure-specific dose thresholds for 20-year cardiovascular risk. Outcome Core Substructure HR per 5 Gy (95% CI) 20-Year Risk Target Dose Threshold (Gy) CAD Aorta 1.59 (1.14–2.21) 1% 13 CAD Aorta 1.59 (1.14–2.21) 2% 20 CAD LAD 1.26 (1.05–1.53) 1% 24 CHF LAD 1.23 (1.02–1.47) 0.5% 21 Valvular Aorta 1.96 (1.15–3.32) 1% 23

Interpretable AI-driven pan-cancer pre-screening of actionable gene fusions from H&E slides: A prototype-guided mixture-of-experts framework to prioritize patients for RNA/DNA sequencing.

e15011 Background: Actionable gene fusions such as ALK, ROS1, and NTRK define critical therapeutic subsets across solid tumors but suffer from the needle-in-a-haystack challenge due to extreme rarity in unselected populations. We hypothesized that those fusion-driven oncogenesis imprints subtle and detectable morphological patterns on routine H&E slides. We present a novel AI framework designed not only to predict fusion status but to enable a cost-effective pre-screening funnel. Methods: We developed PathMoE-Fusion, a prototype-guided mixture-of-experts framework to identify fusion-associated signals directly from routine hematoxylin and eosin (H&E)-stained whole-slide images. The mixture-of-experts architecture enables multiple specialized subnetworks (“experts”) to selectively capture distinct histomorphological patterns, which serve as representative and interpretable prototypes for profiling characteristic phenotypes. PathMoE-Fusion was trained and validated on a pan-cancer cohort of 6,273 patients across 25+ cancer types, including 1,553 fusion-positive cases confirmed by next generation sequencing assays in a CAP/CLIA-certified laboratory (including ALK: 884; RET: 307; ROS1: 255; and NTRK: 107). Model operating points were selected to support conservative pre-screening use, prioritizing high specificity and enrichment of fusion-positive cases. External validation was conducted on independent cohorts, including the TCGA Pan-Cancer Atlas, to assess generalizability across cancer types and domains. Results: PathMoE-Fusion, demonstrated robust pre-screening performance across multiple actionable gene fusion targets, achieving AUROC values ranging from 0.84 (ROS1) to 0.94 (NTRK). Crucially, when deployed as a pre-screening tool under a conservative operating point prioritizing high specificity, the model yielded positive predictive values of 80.2% for ALK and 70.2% for NTRK, indicating marked enrichment of fusion-positive cases relative to their baseline prevalence. In TCGA validation, PathMoE-Fusion showed improved pre-screening performance, with more favorable enrichment characteristics than baseline approaches. Conclusions: PathMoE-Fusion serves as a rapid, low-cost digital enrichment biomarker that transforms the search for rare gene fusions. By prioritizing specificity, it acts as a pre-screening filter to populate a high-yield queue for confirmatory DNA/RNA sequencing. This framework maximizes the cost-effectiveness of molecular diagnostics, ensuring that sequencing resources are concentrated on patients most likely to benefit from life-extending targeted therapies.

Tislelizumab, bevacizumab, and chemotherapy as second-line therapy in PD-L1-positive, RAS-mutant metastatic colorectal cancer: A single-arm, phase II clinical trial.

e15540 Background: The therapeutic options for RAS-mutant metastatic colorectal cancer (mCRC) post first-line therapy are limited, with suboptimal efficacy of standard second-line regimens. Preclinical and clinical data support the synergistic antitumor effect of anti-PD-1 and anti-VEGF agents. Methods: This single-arm phase II study enrolled 42 eligible RAS-mutant, PD-L1-positive mCRC patients who progressed after first-line fluoropyrimidine-based chemotherapy. Patients received tislelizumab, bevacizumab, and switched chemotherapy. The primary endpoint was 6-month progression-free survival (PFS) rate. Results: Between April 2023 and March 2025, 50 patients were screened and 42 patients were eligible. The primary endpoint was met, with a 6-month PFS rate was 66.67% (95% CI:53.83-82.56%). The confirmed ORR was 16.67% (95% CI: 6.97-31.36%), and DCR was 88.10% (95% CI: 74.37-96.02%). The median OS was 16.3 months (95% CI: 13.2-NA months). Treatment-related adverse events occurred in 76.7% of patients, with 4.76% being grade 3–4. Immune-related adverse events were grade 1-2 in 21.43% of patients. No treatment-related deaths occurred. Conclusions: This combination regimen showed promising efficacy and manageable safety, providing a potential therapeutic option for this patient population (ChiCTR2200066728). Clinical trial information: ChiCTR2200066728 .

Cardiac complications as a major driver of ICU utilization in hospitalized cancer patients: A National Inpatient Sample analysis, 2018–2022.

e23167 Background: ICU-level escalation in hospitalized cancer patients is commonly attributed to sepsis or respiratory failure. The relative contribution of acute cardiac complications to critical care utilization remains less well defined. This study evaluated whether cardiac complications independently drive ICU utilization compared with sepsis and respiratory failure during cancer hospitalizations. Methods: A retrospective serial cross-sectional analysis was conducted using adult hospitalizations with a principal diagnosis of malignancy in the 2018–2022 National Inpatient Sample with discharge-level survey weighting. Three competing inpatient drivers were identified using any-diagnosis codes: (1) cardiac complications, including arrhythmia, heart failure, acute myocardial infarction, myocarditis, pericarditis, demand ischemia, stress cardiomyopathy, or cardiac arrest; (2) sepsis; and (3) respiratory failure or acute respiratory distress syndrome. ICU utilization was assessed using a proxy defined as shock or mechanical ventilation. Outcomes included ICU proxy utilization, mechanical ventilation, shock, length of stay, and hospitalization cost. Survey-weighted multivariable models adjusted for demographics, payer, ZIP-code income quartile, admission type, cancer subtype, hospital characteristics, and year, with all three drivers entered simultaneously. Results: Among 961,848 unweighted cancer hospitalizations, representing 4.81 million admissions nationally, cardiac complications occurred in 20.1%, respiratory failure in 10.7%, and sepsis in 3.2%. ICU proxy utilization occurred in 3.49% overall. ICU utilization was lowest in hospitalizations with none of the three drivers (0.84%) and higher in cardiac-only (2.43%), sepsis-only (7.09%), respiratory failure-only (15.03%), and mixed-driver admissions (27.85%). Length of stay and cost increased in parallel, ranging from 5.65 days and $20,230 with no drivers to 14.70 days and $50,397 with mixed drivers. In adjusted models, cardiac complications independently increased odds of ICU utilization (adjusted odds ratio 2.08, 95% CI 2.02–2.14), although respiratory failure (adjusted odds ratio 15.84) and sepsis (adjusted odds ratio 6.22) were stronger per-case predictors. Cardiac complications also independently increased length of stay (+1.43 days) and cost (+$6,169). Conclusions: Cardiac complications are common during cancer hospitalizations and independently contribute to ICU utilization and increased resource use, even after accounting for sepsis and respiratory failure. Although respiratory failure and sepsis confer higher per-case ICU risk, the high prevalence of cardiac complications identifies them as a major driver of critical care burden in oncology hospitalizations.

Endocrine therapy non-adherence and recurrence risk in patients with breast cancer and major depressive disorder.

10595 Background: We evaluated whether non-adherence to oral endocrine therapy explains the increased hazard of recurrence among women with major depressive disorder (MDD) and estrogen receptor-positive breast cancer. Methods: We established a retrospective cohort of veteran women (age ≥18 years), with and without MDD, diagnosed with early-stage invasive estrogen receptor-positive breast cancer between 2010-2019 within the United States Veterans Affairs Healthcare System with follow-up through 2022. We assessed adherence to oral endocrine therapy among continuous users (no treatment gaps ≥ 6 months) over two-years (N = 2,722) and five-years (N = 1,666). Adherence was measured using the proportion of medication days covered and evaluated across thresholds from ≥20%-100%. We used multivariable competing-risks regression to estimate hazard ratios (HR) and 95% confidence intervals (CI) for the statistical interaction between MDD and endocrine therapy adherence on breast cancer recurrence, adjusting for sociodemographic, clinical, and prognostic factors. Results: The hazard of breast cancer recurrence significantly differed between adherent and non-adherent women with MDD beginning at adherence thresholds of ≥60% over two-years and ≥40% over five-years. At these minimal adherence thresholds, breast cancer recurrence was higher for women with MDD who were non-adherent to endocrine therapy over two-years (HR = 1.65; 95% CI = 1.02-2.62) and five-years (HR = 2.10; 95% CI = 1.16-3.79) compared with adherent women without MDD. In contrast, women with MDD who remained adherent over two-years (HR = 1.21; 95% CI = 0.91-1.64) and five-years (HR = 1.38; 95% CI = 0.85-2.22) had hazards of breast cancer recurrence comparable to those of adherent women without MDD. Conclusions: Our findings identify endocrine therapy adherence as a potentially modifiable clinical factor to reduce breast cancer recurrence among women with MDD and estrogen receptor-positive disease.

Global incidence, mortality, and projections of early-onset upper gastrointestinal cancers.

e16339 Background: Cancers of the upper gastrointestinal tract remain major contributors to morbidity and mortality worldwide, yet global patterns in early-onset (EO) disease (age < 50 years) remain poorly characterized. We sought to examine the contemporary global incidence and mortality and analyse expected burden through 2045 using a large population-level database. Methods: GLOBOCAN 2022 was queried for men and women aged 20–49 years with esophageal and gastric cancers. Outcomes were incident cases, deaths, age-standardized incidence and mortality rates (ASR), crude rates, and cumulative risk. Factors were adjusted for populations. Estimates were stratified by continent and HDI. Burden through 2045 was estimated using GLOBOCAN projections, with changes decomposed into population growth/aging and risk effects according to standard GLOBOCAN methodology. Results: In 2022, 80,885 EO gastric cancer cases (ASR 2.4) and 51,259 deaths (ASR 1.6) were estimated globally, while 45,056 EO esophageal cancer cases (ASR 1.4) and 39,196 deaths (ASR 1.2) were projected. The largest share of cases and mortality was in Asia (Table). By HDI, EO gastric cancer incidence (36,170 cases; ASR 2.9) and mortality (21,818 deaths; ASR 1.7) peaked in high-HDI countries, while esophageal cancer incidence and mortality were highest in medium-HDI countries (23,488 cases and 22,734 deaths; both ASR 2.5). From 2022 to 2045, EO esophageal and gastric cancer incidence and mortality are projected to increase across all continents except Europe. Conclusions: EO esophageal and gastric cancers contribute a substantial and uneven global burden, with marked geographic and HDI-related heterogeneity. Incidence and mortality are expected to rise sharply in low- and middle-HDI regions (particularly Africa), while declining in Europe through 2045. Region-specific prevention strategies and health-system preparedness are needed to address the growing burden of early-onset upper gastrointestinal cancers. Incidence and Mortality of Early onset Esophageal and Stomach Cancer (Age 20–49, Both Sexes, 2022-2045). Continent Esophagus Incidence 2022 Esophagus Incidence 2045 %Δ Esophagus Mortality 2022 Esophagus Mortality 2045 %Δ Stomach Incidence 2022 Stomach Incidence 2045 %Δ Stomach Mortality 2022 Stomach Mortality 2045 %Δ Africa 5,979 11,075 +85.2% 5,101 9,446 +85.2% 7,278 13,432 +84.6% 5,828 10,750 +84.5% Asia 33,539 37,242 +11.0% 29,968 33,326 +11.2% 54,869 60,071 +9.5% 33,546 36,764 +9.6% Europe 2,732 2,206 −19.3% 2,011 1,623 −19.3% 7,803 6,265 −19.7% 4,572 3,667 −19.8% Latin America & Caribbean 1,756 2,089 +19.0% 1,383 1,655 +19.7% 7,716 8,682 +12.5% 6,081 6,819 +12.1% Northern America 957 1,131 +18.2% 674 797 +18.2% 2,925 3,366 +15.1% 1,109 1,276 +15.1% Oceania 93 125 +34.4% 59 80 +35.6% 294 385 +31.0% 123 161 +30.9% Global total (sum of regions) 45,056 53,868 +19.6% 39,196 46,927 +19.7% 80,885 92,201 +14.0% 51,259 59,437 +16.0%

An open-label, parallel-cohort, multicenter clinical study of rezvilutamide in combination with androgen deprivation therapy (ADT) with or without docetaxel in the neoadjuvant treatment of oligometastatic prostate cancer.

TPS5148 Background: Oligometastatic hormone-sensitive prostate cancer (omHSPC), defined by ≤5 metastases, represents a distinct disease state with a more favorable prognosis, offering a critical window for treatment intensification aimed at improving long-term outcomes. Cytoreductive radical prostatectomy (cRP) has emerged as a feasible local treatment option for select patients, with phase II data suggesting a survival benefit over systemic therapy alone. The foundation of systemic therapy for metastatic HSPC (mHSPC) is an androgen receptor pathway inhibitor (ARPI) combined with androgen deprivation therapy (ADT). Rezvilutamide, a second-generation ARPI, demonstrated superior efficacy over bicalutamide when combined with ADT in patients with mHSPC. Furthermore, the addition of docetaxel to an ARPI and ADT (triplet therapy) significantly improved survival in mHSPC, including in patients with low metastatic burden. We hypothesize that intensifying neoadjuvant systemic therapy with triplet therapy prior to cRP may maximize tumor cytoreduction and pathologic response in omHSPC. Methods: This is a prospective, open-label, parallel-cohort, multicenter phase II study. Patients with newly diagnosed, biopsy-proven omHSPC (≤5 bone or lymph node metastases on conventional imaging, no visceral metastases) who are candidates for and desire cRP are eligible. All patients undergo baseline prostate-specific membrane antigen positron emission tomography/computed tomography (PSMA-PET/CT) prior to ADT initiation. Patients select their neoadjuvant regimen: Cohort A receives ADT plus oral rezvilutamide (240 mg daily) for 24 weeks; Cohort B receives ADT, rezvilutamide (same dose), plus docetaxel (75 mg/m² IV q3w for 6 cycles) for 24 weeks. Upon completion of the 24-week neoadjuvant therapy, metastases are reconfirmed by conventional imaging, and a second PSMA-PET/CT assessment is performed. All patients subsequently undergo cRP with extended pelvic lymph node dissection within 6 weeks. After surgery, patients may opt to continue with ADT or rezvilutamide at their discretion until disease progression. Based on assumed pCR rates of <10% (Cohort A) and 25% (Cohort B) (α=0.05, power=80%), 40 evaluable patients are required per cohort. Accounting for 20% dropout, 100 total patients (50 per cohort) will be enrolled. Key inclusion criteria: ECOG PS 0-1, adequate organ function. Key exclusion criteria: prior systemic therapy for PCa (except ≤4 weeks of ADT), prior second-generation ARPI use, and prostate pathology results indicating neuroendocrine carcinoma. As of January 15, 2026, enrollment is ongoing with 6 patients accrued to date. No data analysis has been performed. Clinical trial information: ChiCTR2400093262.

Understanding barriers to the clinical breast exam in Telangana, India: A survey study of participants and non-participants in underserved rural and urban areas.

1593 Background: In low-resource settings where mammography is not feasible, Clinical Breast Examination (CBE) performed by trained healthcare workers is an effective alternative for breast cancer early detection. Despite prior success of large-scale CBE programs in India, national uptake remains extremely low. This study evaluated knowledge, attitudes, beliefs, and structural barriers associated with participation in CBE screening among women in rural and urban Telangana, India and assessed factors influencing participation. Methods: We conducted a prospective, nested case-control study in October 2025, in which CBE participants and non-participants were drawn from the same source population. Women aged >35 years who were non-pregnant and able to consent were approached during home visits or at primary health centers (PHCs) to complete standardized surveys; participation in CBE was voluntary. Surveys captured sociodemographic factors, breast cancer knowledge, attitudes toward CBE, stigma, and treatment beliefs. Multivariable logistic regression was used to estimate odds ratios (ORs) and 95% confidence intervals (CIs) for factors associated with CBE participation, adjusting for relevant sociodemographic and belief-based covariates. Results: A total of 669 women were included (urban n=426; rural n=243), of whom 421 participated in CBE (participants) and 248 declined(non-participants). Among 403 CBE participants, 81.1% (n=327) reported never having had a prior breast examination. Belief that CBE is necessary even in the absence of symptoms was strongly associated with participation (81.1% vs 51.6%, p<0.001) as was belief that breast cancer is curable (80.6% vs 56.9%, p<0.001). There was no difference in fear of stigma between participants and non-participants (62.2% vs 65.3%, p=0.61). Fear of losing the breast was highly prevalent among non-participants (72%) and independently associated with lower participation (OR 0.49, 95% CI 0.28–0.87). After adjustment, urban residence was associated with lower odds of participation compared with rural residence (adjusted OR 0.28, 95% CI 0.14–0.56) as was higher household income (>₹100,000 vs ≤₹15,000; adjusted OR 0.24, 95% CI 0.12–0.46). Among non-participants, the most cited facilitators for future participation were examination by a doctor (33.5%), clinic-based exams (28.2%), and convenient timing (28.2%). Conclusions: Participation in CBE is influenced by modifiable belief-based and structural barriers that vary by setting. Addressing fears related to diagnosis and treatment of breast cancer and tailoring CBE delivery strategies may improve uptake and enhance early detection of breast cancer in low-resource settings.

Burden of heart failure–related mortality among U.S. adults with leukemia (1999-2023).

e18623 Background: Cardiovascular disease is the leading non-cancer cause of death among leukemia survivors, with heart failure (HF) accounting for 12.4% of leukemia-related mortality through cardiotoxic treatment effects and inflammatory pathways. Acute HF often develops within one year of diagnosis, compromising therapy. Post–COVID-19 disruptions in cardiac monitoring may further impact survival. We analyzed trends in HF-related mortality among U.S. adults with leukemia from 1999–2023. Methods: Mortality data for adults aged ≥45 years with leukemia and HF from 1999–2023 were obtained from the CDC WONDER database. Age-adjusted mortality rates (AAMRs) per 100,000 and annual percent changes (APC) were estimated using Joinpoint regression, stratified by year, sex, race/ethnicity, census region, and urbanization status. Results: Between 1999 and 2023, 47,156 deaths occurred among leukemia patients with HF. Among adults aged ≥65, AAMRs were stable until 2010 (APC 0.29%, p=0.405) and then increased significantly through 2023 (APC 3.23%, p<0.001). Adults aged 45–64 showed a consistent rise (APC 1.80%, p=0.016). Regionally, the South recorded the highest total deaths (15,381; APC 2.62% from 2010–2020), the West experienced the sharpest short-term increase (APC 5.11%, 2014–2018), and the Northeast showed the greatest overall decline (AAPC −0.13%). Non-Hispanic Whites were the only racial group with a significant overall increase (AAPC 0.76%, p=0.040). Males exhibited a slight increase (AAPC 0.64%), whereas females showed a decrease (AAPC −0.30%). Non-metropolitan areas experienced a larger decrease (AAPC −0.57%) than metropolitan areas (AAPC −0.24%). Conclusions: HF-related mortality among U.S. adults with leukemia has increased in recent years after an initial period of stability or decline. Disparities are notable among males, non-Hispanic Whites, and residents of the Southern U.S. These trends underscore the need for targeted interventions, integrated cardio-oncology care, and improved cardiovascular risk management in leukemia populations. Joinpoint results summary. S. No. Characteristics AAMR (1999-2023) AAPC (95% CI) JP No. 1. Overall 1.7–1.83 0.3(-0.7-1.2) 3 2. Male 2.3–2.66 0.6(-0.4-1.7) 2 3. Female 1.36–1.23 0.01(-0.63-0.64) 1 4. 45-64 yrs 0.12–0.22 1.9(0.7-3.2) 0 5. 65+ years 4.48–4.66 0.2(-0.7-1.1) 3 6. Census Region 3 – South 1.52–1.78 0.5(-1.2-2.3) 3 7. Census Region 4 - West 1.81–1.78 0.7(-0.05-1.5) 1 8. Non Hispanic – White 1.8–2.11 0.6(-0.4-1.8) 3 9. Non Hispanic – Black and African American 1.18–1.24 0.6(-0.5-1.7) 1

Radium-223 without androgen deprivation therapy (ADT) in patients (Pts) with biochemically recurrent prostate cancer (BCR) and PET findings in the bones.

e17107 Background: BCR pts have a rising PSA after definitive surgery or radiation, but negative CT/Tc-99m bone scans. While ADT-based strategies are commonly used, they can be associated with life-altering toxicities. Radium-223 is an alpha-emitting radiopharmaceutical that accumulates in areas of high bone turnover and has demonstrated a survival benefit in metastatic prostate cancer but its benefits as a monotherapy in BCR without ADT have not been explored prior to this study. This strategy may be important as bone metastasis drives morbidity/mortality in prostate cancer. Methods: Eligible pts in this study (NCT04206319) have BCR following definitive therapy, testosterone (T) > 100ng/dL, normal organ and marrow function, and negative CT/Tc-99m bone imaging. Pts must have findings on NaF or PSMA PET suspicious for metastatic bone disease not seen on Tc-99m bone scan. Pts are treated with Radium-223 at the approved dosing of 55 kBq/kg for 6 cycles. PSA declines were defined as 2 or more confirmed declines from an intra-study apex PSA (ISAP; Madan ASCO GU 2018). All pts have pre- and post-treatment PSMA and NaF imaging. Immune responses (primary endpoint) are to be evaluated at study completion. Results: 25 pts have enrolled with 24 currently evaluable for response, with a median age of 69 yrs (57-80) and PSA of 1.75 ng/ml (0.2-49.5). 21/24 pts had PSMA+ scans. 21/24 pts had NaF+ scans. Grade 2 toxicities have been rare and no grade 2 changes in hemoglobin or platelets have occurred. There have been no grade 3/4 toxicities and no cycles missed for toxicity. Two pts with rapid PSA doubling times came off study early for progression and were not evaluable for follow up imaging. 8/24 pts (33%) had confirmed ISAP PSA declines of 12% to 80% lasting from 84 to 682 days, with two ongoing PSA responses. Delayed PSA responses following completion of treatment were observed. 13/21 evaluable pts had evidence of decreased maximum standardized uptake value (SUVmax) on NaF PET imaging after treatment. 5/19 evaluable patients with PSMA PET+ lesions had a radiographic response; two pts had resolution of multiple PSMA+ bone findings, one of whom remains with negative bone findings on PET after 2+ years and with a stable PSA. Conclusions: Radium-223 in BCR is safe and associated with minimal grade 1/2 toxicities. Decreases in SUVmax seen on most NaF scans suggest Radium-223 targeting of suspicious lesions. Improvements in PSMA imaging have been observed. Delayed but confirmed PSA declines have been seen in 33% of pts. Radium-223 monotherapy may have therapeutic activity in BCR and further studies are needed to define its potential role in BCR in patients with PET+ bone findings. Clinical trial information: NCT04206319 .

Machine learning approach to identify country-specific drivers of global cancer outcomes.

1596 Background: Global disparities in access to cancer diagnostics and care drive substantial variation in cancer outcomes worldwide. Identifying actionable health system factors at the country level is critical to improving survival and closing global equity gaps. We applied explainable machine learning methods to explore the interrelations among these interrelated factors. Methods: We developed and validated a machine learning model using data from 185 countries, obtained from GLOBOCAN 2022, WHO, and the World Bank. Mortality-to-incidence ratios (MIRs) were predicted using repeated leave-one-country-out cross-validation. SHapley Additive exPlanations (SHAP) analysis decomposed each prediction into country-specific health system-feature attributions, revealing the principal drivers of cancer outcomes. Results: The model achieved robust performance (R²=0.852; RMSE=0.057; Pearson r=0.923, p<0.001). Globally, GDP per capita, radiotherapy center density, and the universal health coverage (UHC) index were the top contributors to MIRs. SHAP analysis revealed significant heterogeneity, identifying distinct priority drivers for each country. For example, radiotherapy infrastructure was most impactful in Turkey, UHC in Brazil and Ghana, and GDP per capita in Malaysia and China. Higher health spending as a percent of GDP was often paradoxically associated with higher MIR, underscoring the need not only for sufficient funding but also for strategic allocation. We developed a web tool that provides policymakers with country-specific SHAP estimates. Conclusions: Explainable machine learning translates global associations into actionable, country-specific insights. Access to radiotherapy, workforce development, and UHC expansion are consistently associated with improved cancer survival. These findings support resource prioritization in national cancer control planning and inform hypothesis generation for future causal studies. National-level analyses guide targeted investment in infrastructure and health coverage, potentially accelerating progress toward reducing global disparities in cancer mortality.

The Remote Emotional Support and Symptom Self-Management Intervention (REMI): A pilot study.

e24070 Background: Symptom burdens related to advanced cancer remain high. The Remote Emotional Support and Symptom Self-Management Intervention (REMI) incorporates palliative care and Acceptance and Commitment Therapy principles to encourage patients living with metastatic disease to engage in symptom self-management activities. REMI is an AI-driven, supportive care program that presents empathic messaging, utilizes the conversational aspects of Amazon ALEXA, and offers symptom management recommendations. The purpose of this study is to assess the acceptability of generated messages that would be part of the REMI device through participant evaluation. Methods: Patients with metastatic solid tumors (n = 20 participants) were recruited from the UPMC Hillman Cancer Center network, having at least 20% of the patient sample categorized as rural. Empathic messages generated by the ChatGPT-4o model were evaluated by participants through both quantitative evaluation packet surveys and semi-structured interviews. Evaluation questions were created by our expert team, pulling from multiple validated surveys exploring symptom management and technology acceptance. Each quantitative question was scored on a 1-10 scale with higher scores translating to higher acceptability. Results: There was overall acceptability of these messages, as participants reported high average scores for all quantitative evaluation questions. Table 1 describes the summary statistics for the quantitative portion of this study. There were varying responses regarding language use; some participants enjoyed personable language while others found it condescending. The majority of participants (n = 16, 80%) expressed interest in using a conversational device based on the information given during this pilot. Conclusions: This pilot project demonstrates early acceptability of the use of AI-generated messages as a part of the REMI program. Future work will focus on usability testing of the REMI device with these patient-informed, empathic messages to help create an effective symptom management program for those living with metastatic disease. Summary statistics for quantitative questions. Average (sd) Message clarity 8.65 (1.79) Tone of message 8.41 (1.88) Message relatability 7.45 (2.50) Message believability 7.82 (2.23) To what extent do you feel this message can help patients living with cancer manage their symptoms? 7.72 (2.36) How likely would you be to change your behavior based on this message? 6.72 (2.88)

Risk-adapted step-up local therapy for malignant portal vein tumor thrombus from hepatic metastases: A real-world study of microwave ablation with and without iodine-125 seed implantation.

4235 Background: Malignant portal vein tumor thrombus (PVTT) arising from hepatic metastases is associated with poor prognosis, rapid hepatic decompensation, and limited effective local treatment options despite systemic therapy. In real-world practice, local treatment intensity is frequently adapted to disease burden and treatment response. We evaluated outcomes of a risk-adapted, step-up local treatment strategy using microwave ablation (MWA) alone or in combination with iodine-125 (I-125) seed implantation for patients with malignant PVTT. Methods: We retrospectively reviewed patients with hepatic metastases treated with MWA with or without I-125 seed implantation between 2023 and July 2025. In clinical practice, MWA alone was preferentially used for patients with relatively limited disease burden, while MWA plus I-125 was employed as an escalation strategy for patients with more advanced PVTT or disease progression following MWA alone. The primary endpoint was PVTT-directed local outcome, defined as portal vein patency and/or local control of tumor thrombus on follow-up imaging. Secondary endpoints included progression-free survival (PFS), overall survival (OS), hepatic function changes, and treatment-related toxicity. Survival analyses were considered exploratory. Results: A total of 73 patients were included, of whom 50 were treated with MWA alone and 23 with MWA plus I-125 seed implantation. Patients receiving MWA plus I-125 demonstrated more advanced disease characteristics, including universal presence of PVTT and higher disease stage, consistent with treatment escalation by disease severity. Among patients with PVTT, MWA plus I-125 achieved portal vein patency and/or local control in 65% at first follow-up imaging. Treatment-related toxicities were manageable, with no unexpected grade ≥3 adverse events attributable to I-125 implantation. Median PFS was 11.5 months in patients treated with MWA alone and 9.9 months in patients treated with MWA plus I-125, consistent with greater baseline disease burden in the latter group. Conclusions: MWA alone and MWA plus I-125 function as complementary components of a risk-adapted, step-up local treatment strategy for malignant PVTT from hepatic metastases. Escalation to MWA plus I-125 appears feasible in patients with advanced or refractory PVTT and demonstrates a potential signal for improved PVTT-directed local control and portal vein patency. Prospective studies focused on PVTT-specific endpoints are warranted.

Self‐Limiting Polymerization‐Induced Crystallization‐Driven Self‐Assembly (SL‐PI‐CDSA) Enables Templated Synthesis of Chiral Plasmonic, Hybrid 2D Hexagonal Assemblies

ABSTRACT Examples of self‐regulating synthetic self‐assembly are relatively few, with most known chemical systems relying on kinetic rather than thermodynamic control. Herein, we demonstrate the rapid generation ( t ≤ 5 min) of size‐tunable ultralow dispersity ( Ð ≤ 1.01) 2D hexagonal nanosheets governed by self‐limiting self‐assembly (SLSA). Self‐assembly in natural systems occurs with exquisite control of structure, function, and dimension. We demonstrate that key aspects of biological assembly can be rationally applied toward the development of bottom‐up approaches for the construction of chiral nanomaterials. To this end, self‐limiting polymerization‐induced crystallization‐driven self‐assembly (SL‐PI‐CDSA) of modular and templating aryl isocyanide (AIC) monomers yields functional 2D assemblies permitting post‐polymerization/assembly modifications. Detailed study of the internal and external structure of the hexagonal nanosheets reveals topological defects which offer mechanistic insights into both their assembly and subsequent utilization. Specifically, these features enable fabrication of chiral hybrid organic–inorganic nanomaterials incorporating chiral plasmonic metal nanoparticles (MNPs). Our results suggest that the synergistic interplay of template‐driven confinement and hierarchical chirality induce symmetry breaking of in situ‐generated gold MNPs. We anticipate that the platform presented will facilitate fabrication of new hybrid, chiral organic–inorganic nanostructures.

Precision nanomedicines in glioblastoma therapy and challenges remain

Molybdenum‐Pocket Driven Low‐Platinum Oxygen Reduction Catalysts for 100‐Watt‐Scale Fuel Cell Stacks

ABSTRACT Developing highly efficient and low‐platinum (Pt) proton exchange membrane fuel cell (PEMFC) stacks is imperative for their commercialization. However, ultralow‐loading Pt catalysts (<0.10 mg cm −2 ) are inherently less active and unstable due to high oxygen resistance, particularly under practical stack operating conditions. Here, we present a molybdenum oxide‐pocket‐driven Pt 2 Co (MoO 3 ‐Pt 2 Co) alloy to tackle the aforementioned challenge, where MoO 3 with abundant oxygen vacancies can act as the pivotal “oxygen storage pocket” to boost the oxygen reduction reaction (ORR) activity and minimize the leaching of Co. Consequently, the MoO 3 ‐Pt 2 Co/C‐based membrane electrode assembly (MEA) enables exceptional peak power densities of 3.20 W cm −2 and 1.73 W cm −2 in H 2 ‐O 2 and H 2 ‐air, respectively, with a low Pt loading of 0.10 mg cm −2 , outperforming cutting‐edge MEAs. Meanwhile, the MoO 3 ‐Pt 2 Co/C‐based MEA can retain a record‐breaking mass activity of 1.68 A mg −1 after 30k‐cycle accelerated stress tests and can be operated stably at 0.65 V beyond 550 h. Most importantly, we develop a MoO 3 ‐Pt 2 Co/C‐based fuel cell stack that delivers an excellent rated power of 123 W in H 2 ‐air, which can project Pt utilization of 0.0975 g Pt kW −1 for a 100‐kW hydrogen fuel cell vehicle, exceeding the US Department of energy (DOE) ultimate target of 0.10 g Pt kW −1 .

Ultrabright, Efficient, and Stable Green Quantum Dot Light‐Emitting Diodes Through Effective Heat Dissipation

ABSTRACT Heat generation induced by electrical driving is primarily responsible for the device failure in quantum dot light‐emitting diodes (QD‐LEDs). Elucidating the thermally induced degeneration mechanism is crucial to enhancing operational lifetime and luminescent efficiency of the devices. Heightening heat dissipation can mitigate device degradation by utilizing high thermal conductivity materials, allowing for ultrahigh luminance and enhanced efficiency at high driving voltages. Here we constructively propose an underlying mechanism by which heat accumulation within the device enhances carrier delocalization in QDs, and induces the accumulation of numerous holes and electrons in non‐recombination regions, which reduce the charge density available for recombination, resulting in restricted peak brightness and device failure. On this basis, a universally feasible encapsulation strategy is developed to accelerate heat dissipation, enabling record‐breaking green QD‐LEDs with a luminance of 2 036 000 cd m −2 , an external quantum efficiency of 32.1%, and a T 95 operation lifetime (time for the luminance decreasing by 95%) of more than 32 000 h at 1000 cd m −2 .

Combined therapy using gemcitabine and cisplatin chemotherapy, lenvatinib, and adebrelimab for patients with advanced and unresectable intrahepatic cholangiocarcinoma: A prospective, single-arm, phase II trial.

e16190 Background: The prognosis of advanced intrahepatic cholangiocarcinoma (ICC) remains unsatisfactory with current gemcitabine-based chemotherapy in combination with PD-1/PD-L1 inhibitors. This study aimed to investigate the efficacy and safety of aderbrelimab (PD-L1 inhibitor) and lenvatinib in combination with gemcitabine and cisplatin as first-line treatment in advanced and unresectable ICC patients. Methods: This prospective, single-arm, multicenter, phase II study included treatment-naive adult (≥18 years) patients with unresectable and advanced ICC. All patients received 1200mg aderbrelimab Q3W via intravenous (IV) infusion, 8mg lenvatinib QD orally, and 1g/m² gemcitabine plus 25 mg/m² cisplatin on Day 1 and Day 8, Q3W by IV for 6 cycles, followed by maintenance aderbrelimab and lenvatinib until disease progression (PD) or unacceptable toxicity. The primary endpoint was objective response rate (ORR) per Response Evaluation Criteria in Solid Tumors 1.1 (RECIST 1.1). The secondary endpoints included overall survival (OS), progression-free survival (PFS), disease control rate (DCR), and safety. Results: Between April 2024 and November 2025, 26 pathologically-confirmed advanced ICC patients with a median age of 60.5 (range, 45-78) years, including 20 men (76.9%), were enrolled at five centers. At the end of last follow-up (January 20, 2026), the ORR was 46. 2% (12/26; 95% CI: 26.6%-66.6%), and disease control rate (DCR) was 92.3% (24/26; 95% CI:74.9%-99.1%). Median follow-up time was 8.1 months. 12 patients experienced disease progression and 4 patients have died. The median PFS was 8.5 months. The median OS have not been reached. The most common grade 3 or 4 treatment-related adverse events (TRAEs) were neutropenia and thrombocytopenia observed in 6 (23.1%) and 5 (19.2%) patients, respectively. No unacceptable toxicity or treatment-related deaths occurred. Conclusions: Aderbrelimab and lenvatinib in combination with gemcitabine and cisplatin showed promising efficacy with reasonable tolerability in advanced ICC patients. Survival data is immature and will be updated in the future. Clinical trial information: NCT06298968 .