Browse Articles

Divergence between progression-free and overall survival treatment effect sizes in antibody-drug conjugate trials: A meta-epidemiologic study.

e13076 Background: Progression-free survival (PFS) is frequently used as a primary endpoint in contemporary oncology trials, including antibody–drug conjugate (ADC) studies in metastatic breast cancer (MBC) settings, despite uncertainty regarding its validity as a surrogate for overall survival (OS). A meta-epidemiological analysis was performed to quantify the systematic differences between the treatment effects on PFS and OS in ADC randomized clinical trials (RCTs). Methods: Phase II–III ADC RCTs with MBC patients reporting hazard ratios (HRs) with 95% confidence intervals (CIs) for both progression-free survival (PFS) and overall survival (OS) were included. Differences in treatment effects within each trial were quantified using the ratio of hazard ratios (rHR = HR_PFS/HR_OS). Random-effects models were used to pool the rHRs across the trials. Prespecified subgroup and meta-regression analyses were used to evaluate the trial characteristics, clinical context, and regulatory status. Sensitivity analyses assessed the robustness across varying assumptions of the within-trial correlation between the endpoints. Results: Twenty-one RCTs evaluating nine ADCs in patients with MBC were included, comprising three phase II (14%) and 18 phase III (86%) studies. The pooled rHR demonstrated significantly larger treatment effects for PFS than for OS (rHR, 0.79; 95% CI, 0.73–0.86), with no between-trial heterogeneity (I² = 0%, τ² = 0). Sensitivity analyses assuming within-trial correlations between PFS and OS ranging from 0 to 0.95 yielded consistent results. Greater divergence between endpoints was observed in trials enrolling patients with triple-negative breast cancer (rHR, 0.72; 95% CI, 0.61–0.85) and HER2-low disease (rHR, 0.77; 95% CI, 0.62–0.96). Phase III trials showed larger differences than phase II trials (rHR, 0.79 vs. 0.89). When OS was a secondary endpoint, divergence was greater than when OS was primary (rHR, 0.77 vs. 0.87). Across trials, 16 of 21 studies (76%) demonstrated concordant effect directions between PFS and OS: 11 (52%) with a statistically significant benefit for both endpoints and 5 (24%) with neither endpoint significant. Discordance occurred in five trials (24%), all characterized by a statistically significant PFS benefit without a corresponding OS improvement. No trial demonstrated an OS benefit in the absence of a PFS benefit. Conclusions: Across ADC trials, PFS systematically overestimated OS treatment effects, underscoring the limitations of PFS as a standalone surrogate endpoint. These findings support the cautious interpretation of PFS-driven efficacy and reinforce the need for endpoint validation frameworks tailored to the mechanism, disease context, and trial design.

Identifying exposome-driven cancer incidence patterns using multi-cancer principal components analysis.

e22606 Background: Populations are exposed to complex mixtures of environmental, behavioral, and social factors that influence multiple cancers simultaneously. While exposomic approaches have increasingly been applied to individual cancers, multicancer analyses leveraging shared, geographically patterned exposures remain underutilized. As many exposures cluster spatially and affect multiple malignancies, cancer incidence may reflect shared regional risk structures. Methods: Age-adjusted average incidence rates for 12 cancers from 2017–2021 were obtained from the NCI State Cancer Profiles. Counties with complete, non-suppressed data were included (n = 1567). Principal component analysis (PCA) was applied to the county-by-cancer incidence matrix to identify orthogonal multicancer incidence patterns. County-level exposures from national surveillance datasets were correlated with each principal component to characterize exposome signatures. Results: PCA identified 12 orthogonal components of multicancer incidence, with the first three explaining the majority of shared variation across counties. PC1 represented a general cancer burden axis dominated by smoking and behavior-associated malignancies. PC2 reflected cancer screening and healthcare access, distinguishing screen-detected from screen-preventable cancers.PC3 captured a metabolic–industrial exposure axis and showed the strongest association with recent county-level changes in cancer incidence (r = 0.59). Conclusions: Population-level cancer burden appears structured along a small number of orthogonal axes spanning healthcare access and screening, health behaviors, and environmental exposures. This exposome-informed, multicancer framework uses PCA to capture geographically patterned influences on cancer incidence and provides a scalable strategy for prioritizing exposures and markers for downstream causal and mechanistic investigation. Multicancer principal components and associated exposome signatures. PC PC Variance (%) Dominant Cancers (r) Key Correlated Exposures (r) Conceptual Axis PC1 25.2 Lung (0.40)Colon (0.34)Oral (0.34)Bladder (0.33) Carbon tetrachloride (0.37)Smoking (0.37)Obesity (0.30) General cancer burden PC2 16.4 Melanoma (0.45)Breast (0.41)Colon (−0.38) Median household income (0.60)Colonoscopy screening (0.53)Social Vulnerability Index (−0.52) Screening / access PC3 10.1 Prostate (0.63)Breast (0.36)Pancreatic (0.33) Diabetes (0.28)PM2.5 (0.24)Ethylene oxide (0.18) Metabolic–industrial Datasets: EPA National Air Toxics Assessment (NATA, 2025); US Geological Survey EPest (2013–2017); BRFSS (2018–2020); USDA Economic Research Service (2020); U.S. Census (2023); NCI State Cancer Profiles (2017–2021). Cancers in study: Bladder, Breast, Colon, Kidney, Leukemia, Lung, Lymphoma, Melanoma, Oral, Pancreatic, Prostate and Uterine.

Patient-reported global health to predict negative health outcomes in patients with advanced sarcoma.

11554 Background: Many patients with sarcoma deal with adverse impacts on their (QoL) related to their underlying disease as well as treatment-related effects. Unplanned hospital admissions and emergency department (ED) visits occur from uncontrolled symptoms that delay treatment and lead to poor hospital system metrics. The PROMIS Global Health v1.2 (PROMIS-10) survey is a validated tool to assess patients’ physical and mental well-being at the time of the survey, with a higher number indicating a better state of health. There are limited studies evaluating longitudinal QoL assessments during sarcoma treatment. Our objective was to analyze PROMIS-10 scores for patients with advanced sarcoma and their temporal relationship with negative outcomes of interest. Methods: We performed a retrospective registry-based longitudinal analysis of routinely collected PROMIS-10 surveys at The Ohio State University. Patients were consented to the sarcoma registry (NCT02677961) from 6/1/2018 to 12/31/2024. We collected baseline, serial and overall change in physical and mental PROMIS-10 raw and normalized t-scores for each patient during their treatment as well as clinical characteristics, mortality, disease progression (RECIST), and ED visits/unplanned hospital admissions. Survival analyses with time-varying covariates were performed using the marginal Cox model approach for statistical analysis. Results: A total of 241 patients were included in the study. Patients with higher PROMIS-10 physical scores had a 13% decrease in risk of death (HR 0.87; P < 0.001), 3% decrease in risk of progression (HR 0.97; P = 0.027), and 6% decrease in risk of ED visit/unplanned hospital admission (HR 0.94; P < 0.001). Patients with a higher PROMIS-10 mental health score had a 7% decrease in risk of death (HR 0.93; P < 0.001), but no association with disease progression or risk of hospital utilization. The larger positive change in physical PROMIS-10 score from baseline to final assessment was associated with an 11% decreased risk of mortality (HR 0.89; P < 0.001) and 9% decreased risk of ED visit/hospital admission (HR 0.91; P = 0.032), but no association with progression. The larger positive change in mental PROMIS-10 score was associated with a 7% decrease in risk of progression (HR 0.93; P = 0.010) but there was no association with mortality or hospital utilization. Conclusions: We demonstrated that obtaining longitudinal patient-reported outcomes to evaluate patients’ physical and mental health during sarcoma treatment can predict mortality, disease progression and ED visits/unplanned hospital admissions. Higher physical PROMIS-10 scores are strongly associated with decreased mortality and hospital utilization risk, whereas a larger positive change in mental PROMIS-10 scores is associated with decreased risk of progression. To our knowledge, this is the largest cohort (241 patients) of advanced sarcoma analyzing QoL impacts.

Differential effects of Acuity Circles policy on waitlist outcomes in hepatocellular carcinoma transplantation by region, insurance status, and ethnicity.

e23251 Background: The Acuity Circles (AC) policy replaced Donation Service Areas with distance-based sharing to mitigate geographic inequities in Feb 2020. However, concerns persisted that the logistics of broader sharing might disproportionately disadvantage patients with lower socioeconomic status (SES). We investigated whether the intended geographic equalization inadvertently exacerbated disparities among racial minorities and insurance groups. Methods: Using UNOS/OPTN data, we retrospectively analyzed a total of 24,464 adult hepatocellular (HCC) candidates listed during symmetric 21-month periods before and after AC implementation. Time to deceased donor liver transplantation (DDLT) was assessed using adjusted restricted mean failure time (RMFT). Difference-in-differences models were used to estimate differential policy effects across race/ethnicity and insurance status. Results: The policy successfully smoothed geographic extremes: wait times in New York/eastern VT decreased by 58.1 days (95% CI [-77.1, -39.1]; p < 0.001), while those in the Southeast increased by 71.5 days (95% CI [54.2, 88.9]; p < 0.001). Contrary to SES concerns, the policy acted as an economic equalizer, narrowing the waitlist gap for public insurance holders by 12.3 days (95% CI [-16.1, -8.6]; p < 0.001).However, racial impact was heterogeneous. While Asian candidates saw a reduced disparity of 31.8 days (95% CI [-49.8, -13.9]; p < 0.001), the pre-existing gap for Hispanic patients widened by an additional 18.8 days post-policy (95% CI [11.9, 25.8]; p < 0.001). Notably, Texas/Oklahoma also experienced a significant waitlist time deterioration of 34.7 days (95% CI [23.2, 46.2]; p < 0.001). Conclusions: AC succeeded in bridging geographic and economic divides, benefiting public insurance and Asian candidates. However, it paradoxically worsened outcomes for Hispanic candidates. This suggests that heterogeneous geographic shifts - specifically waitlist deterioration in regions with high Hispanic density such as Texas/Oklahoma - may have structurally transferred disadvantages to specific racial groups. Future policy recalibrations must address these geo-racial interactions to ensure equitable access.

Impact of aryl hydrocarbon receptor (AHR) modulation on survival in pre-clinical head and neck models via microbial tryptophan pathway.

e14594 Background: Metastatic head and neck squamous cell carcinoma (HNSCC) remains associated with poor outcomes. We have previously demonstrated that microbial tryptophan metabolites are associated with resistance to immune checkpoint blockade, likely through modulation of aryl hydrocarbon receptor (AhR) signaling. Based on this rationale, we evaluated the efficacy of a novel AhR-modulating combination regimen to improve progression-free and overall survival. Methods: Using a syngeneic HNSCC mouse model implanted with MOC1 cells, ten mice were treated with a novel AhR-modulating combination (JS0001). Tumor volume was measured twice weekly until humane endpoint. Mice treated with standard-of-care anti-mouse programmed cell death-1 (anti-PD-1) antibody served as controls. Results: Treatment with JS0001 resulted in a significant reduction in tumor growth rate and improved progression-free survival compared with controls (tumor growth slope estimate −0.87, SE 0.02; p < 0.00001). To investigate mechanisms underlying differential tumor response, systemic and tumor metabolomic profiling was performed. Treatment was associated with significant systemic reductions in multiple microbial tryptophan metabolites, with the strongest effect observed for indole-3-pyruvic acid (IPyA; percent change per day −1.3%, adjusted p < 0.002). Concordant intratumoral reductions in IPyA and indole acetaldehyde (IAAld) were observed in treated mice (estimates 0.70 and 0.47, respectively; p < 0.05). Conclusions: These data support AhR modulation as a strategy to enhance antitumor efficacy and provide preclinical justification for evaluating this novel combination in early-phase clinical trials.

Outcomes of surgical resection combined with intraoperative cesium-131 brachytherapy in brain metastases: A systematic review and meta-analysis.

e14014 Background: Local control after surgical resection of brain metastases remains difficult to achieve. Although postoperative stereotactic radiosurgery (SRS) and whole-brain radiotherapy (WBRT) improve intracranial control, both approaches can delay treatment and increase the risk of radiation necrosis and cognitive decline, particularly in patients with large or rapidly growing tumors. Intraoperative cesium-131 (Cs-131) brachytherapy allows immediate, targeted radiation to the resection cavity and may help address these challenges. However, available evidence is limited, and outcomes have not been well quantified. Methods: We performed a systematic review and single-arm meta-analysis following PRISMA guidelines to evaluate outcomes after surgical resection with intraoperative Cs-131 brachytherapy in patients with recurrent or high-risk newly diagnosed brain metastases. Event rates and standard errors were extracted and pooled using a random-effects model. Heterogeneity was assessed using the I² statistic, and sensitivity analyses were conducted by excluding outlier studies. Primary outcomes included local freedom from progression (LFFP), radiation necrosis, and surgical complications. Results: Nine studies including 479 patients were analyzed. The pooled LFFP rate was 0.61 (95% CI, 0.29–0.94; p = 0.0002), with considerable heterogeneity (I² = 100%). After excluding three outlier studies, LFFP increased to 0.93 (95% CI, 0.91–0.96; p < 0.00001), with no remaining heterogeneity. Radiation necrosis was reported in five studies (n = 296) and occurred in 11% of patients (95% CI, 0.05–0.18; p = 0.001), decreasing to 9% (95% CI, 0.03–0.14; p = 0.01) on sensitivity analysis. Surgical complications were observed in 11% of patients (4 studies; 95% CI, 0.08–0.13; p < 0.00001; I² = 23%). Conclusions: Surgical resection with intraoperative Cs-131 brachytherapy is associated with high rates of local control and a low incidence of treatment-related toxicity in patients with brain metastases. When adjusted for outliers, local control rates approach 93%, comparing favorably with published outcomes for postoperative SRS. Importantly, immediate radiation delivery may reduce treatment delays without increasing surgical risk. While these findings are encouraging, they are based on small, heterogeneous retrospective studies. Prospective trials are needed to better define the role of Cs-131 brachytherapy alongside standard postoperative radiation approaches.

Weekend admission and early in-hospital mortality among patients with hematologic malignancies admitted for febrile neutropenia.

11087 Background: Weekend admissions have been associated with worse health outcomes compared with weekday admissions, a phenomenon known as the ‘weekend effect’. While predictors of in-hospital outcomes among patients with febrile neutropenia (FN) and hematologic malignancies are well described, the impact of the ‘weekend effect’ in this population remains unclear. Our study assessed the association between weekend admission, 3- and 7-day mortality. We further evaluated whether septic shock mediates the relationship between weekend admission and inpatient mortality, regardless of length of inpatient stay. Methods: We conducted a retrospective analysis of the National Inpatient Sample (2016 to 2020). Using ICD-10 codes, we identified patients with hematologic malignancies admitted for FN (D70.X, R50.81 and R50.9). Weekend admission was the primary exposure, while 3- and 7-day in-hospital mortality were the outcomes assessed. Generalized linear models were used to assess the association between weekend admission and each outcome, adjusting for sociodemographic characteristics, Charlson Comorbidity Index, mechanical ventilation, acute renal failure, protein energy malnutrition, cancer type, and length of stay. We performed a mediation analysis to assess whether septic shock mediated the association between weekend admission and overall mortality, estimating natural direct, indirect, and marginal total effects. Results: Among 45,540 patients in this study, 57% were male, 65% were white (mean age was 51 years). Patients had lymphoma (27%), leukemia (54%), multiple myeloma (10%), and multiple/other hematologic malignancies (9%). Weekend admission was associated with a higher risk of 3-day (adjusted risk ratio [adj RR] 1.48; 95% CI: 1.21–1.83) and 7-day mortality (adj RR 1.32; 95% CI: 1.12–1.56). When outcomes were assessed across the entire inpatient stay, weekend admission was associated with higher risk of septic shock (adj RR 1.14; 95% CI, 1.05–1.25). Further mediation analysis demonstrated that septic shock significantly mediated the association between weekend admission and overall inpatient mortality – indirect effect (adj RR 1.03; 95% CI 1.01–1.04), whereas the direct effect (adj RR 1.06; 95% CI 0.94–1.20) and overall effect (adj RR 1.09; 95% CI: 0.96–1.23) of weekend admission on mortality were not statistically significant. Conclusions: Weekend admission was associated with a higher risk of 3- and 7-day mortality. Septic shock partially mediated the relationship between weekend admission and mortality when assessed across the entire inpatient stay. The absence of a direct weekend effect suggests that health system-level factors contributing to septic shock such as delays in recognition, escalation, or time to antibiotics, may represent key intervention targets to improve mortality outcomes in this high-risk population.

Influence of neoadjuvant spatially fractionated stereotactic body radiotherapy on antitumor immunity and pathological response to toripalimab-based immunochemotherapy in resectable stage IIA-IIIB non-small cell lung cancer.

8046 Background: Neoadjuvant immunochemotherapy has improved outcomes in resectable non–small cell lung cancer (NSCLC); however, pathological complete response (pCR) rates remain limited, and strategies to further amplify antitumor immunity are needed. Spatially fractionated stereotactic body radiotherapy (SF-SBRT, Lattice) delivers spatially separated high-dose regions "peaks" to promote immunogenic cell death while maintaining low-dose regions "valleys" that preserve the tumor immune microenvironment (TIME). We hypothesized that integrating SF-SBRT with toripalimab-based immunochemotherapy could reprogram the TIME and enhance pathological response in resectable NSCLC. Methods: This prospective phase II study enrolled resectable stage IIA–IIIB NSCLC (NCT06293690). Patients received one fraction of Lattice SF-SBRT (12 Gy peak / <4 Gy valley) on Day 1, followed by toripalimab plus chemotherapy on Day 2, administered every 3 weeks for 2 cycles prior to surgery. The primary endpoint was major pathological response (MPR). Secondary endpoints included pCR, radiological response, nodal downstaging, event-free survival, overall survival, and safety. Single-cell RNA sequencing (scRNA-seq) was performed to characterize treatment-associated immune remodeling. Results: 20 patients completed neoadjuvant therapy and underwent curative-intent resection, R0 resection was achieved in all patients without treatment-related surgical delays. MPR was observed in 95.0% (19/20) of patients, including a pCR rate of 55.0% (11/20). The radiological objective response rate was 100.0%. Nodal downstaging to ypN0 occurred in 82.4% (14/17) of patients with baseline nodal involvement, corresponding to an overall ypN0 rate of 85.0%. Grade 3–4 treatment-related adverse events occurred in 15.0% (3/20), with no grade ≥2 radiation pneumonitis. scRNA-seq revealed enhanced cytotoxic T-cell activation, reduced immunosuppressive myeloid populations, and coordinated immune remodeling following neoadjuvant treatment. Conclusions: Neoadjuvant SF-SBRT combined with toripalimab-based immunochemotherapy was feasible, safe, and achieved high pathological response rates in resectable stage II–III NSCLC. These findings suggest that SF-SBRT may function as an immune-reprogramming strategy that amplifies neoadjuvant immunotherapy efficacy and warrants further validation in randomized clinical trials. Clinical trial information: NCT06293690 . Characteristics No. of Patients (%) Radiological Response (RECIST 1.1) CR 2 (10.0) PR 18 (90.0) Pathological Response Major pathological response (MPR)* 19 (95.0) └ Pathological complete response (pCR) † 11 (55.0) ypN0 ‡ 17 (85.0) * MPR: < 10% residual viable tumor (RVT); † pCR: 0% RVT in primary & nodes; ‡ ypN0: 14/17 patients with baseline N + achieved downstaging.

A quality improvement initiative to improve genomic testing rates in patients with prostate cancer.

e23316 Background: The therapeutic landscape of metastatic prostate cancer has rapidly evolved with the integration of precision oncology. Emerging data from the AMPLITUDE trial demonstrated that adding niraparib to abiraterone significantly improves rPFS in patients with BRCA-mutated mCSPC, leading to FDA approval in December 2025. These findings, together with the established role of PARP inhibitors in metastatic mCRPC, highlight the importance of universal somatic testing at the time of metastatic prostate cancer (mPC) diagnosis to guide long-term treatment strategies. We identified variability in institutional genomic testing rates at Miami Cancer Institute (MCI) among patients with prostate cancer and initiated a quality improvement (QI) intervention to improve genomic testing rates. Methods: We initiated a multidisciplinary QI project involving stakeholders from medical oncology, radiation oncology, urologic oncology, and the department of genetics. The primary intervention consisted of a CME-accredited, provider-focused educational program. Clinical knowledge was assessed using a 10-point structured questionnaire developed by institutional genitourinary leadership, in accordance with NCCN and ASCO guidelines, to measure pre- and post-intervention proficiency in genomic testing. In parallel, bilingual patient educational resources were deployed to enhance shared decision-making and patient health literacy. Additionally, institutional practice at the MCI was standardized to implement universal germline testing for all newly diagnosed prostate cancer patients and universal somatic testing for all patients with mPC. Results: Pre-intervention data included 647 patients with metastatic prostate cancer (54.1% Hispanic). Overall, 35.5% underwent somatic genomic testing, with lower utilization among Hispanic vs non-Hispanic patients (31.4% vs 40.1%; χ² = 6.7, p = 0.01). Twenty-eight providers completed the pre-intervention assessment and 21 completed post-intervention. Knowledge of genomic testing improved significantly after the CME intervention (mean Δ 1.38; p = 0.0002), with the largest gains among Surgical Oncology APPs (Δ +1.5) and Radiation Oncology physicians (Δ +1.4). Medical Oncology physicians achieved the highest post-intervention scores (mean 9.3/10). Conclusions: This QI initiative identified a significant ethnic disparity in somatic genomic testing utilization among Hispanic and non-Hispanic patients with mPC at MCI. A CME-based, multidisciplinary educational intervention was associated with improved provider knowledge regarding guideline-concordant genomic testing. Ongoing implementation of universal genomic testing pathways, along with continued patient education, represents a feasible strategy to improve testing uptake and promote equitable access to precision therapies in a diverse patient population served at MCI.

WINSHIP 4468-18: Phase II evaluation of the effect of 2- versus 6-hour oxaliplatin infusions on neuropathy and pharmacokinetics in patients with gastrointestinal cancers.

TPS12174 Background: Oxaliplatin, a third-generation platinum agent, improves outcomes in gastrointestinal (GI) cancers when combined with fluoropyrimidines. However, it is also associated with clinically significant acute and chronic neurotoxicity. Approximately 33% of patients develop grade 3–4 sensory neuropathy, and up to 11% discontinue therapy due to neuropathy. Acute neurotoxicity is linked to the maximum plasma concentration (Cmax) of platinum ultrafiltrate, and greater acute sensory neuropathic symptoms predict later chronic neuropathy. Prior data suggest that extending infusion duration from 2 hours to 4–6 hours lowers Cmax by ~32–50% and reduces recurrence of acute neuropathic symptoms. We hypothesize that a 6-hour oxaliplatin infusion will reduce platinum ultrafiltrate Cmax and decrease the incidence and severity of chronic sensory neuropathy. Methods: WINSHIP 4468-18 is a randomized, open-label phase II trial enrolling a real-world population of patients with gastrointestinal (GI) cancers for whom oxaliplatin-containing combination therapy is indicated. All participants receive IV infusion of oxaliplatin 85 mg/m² every 2 weeks as part of a standard combination regimen. Sixty patients will be randomized 2:1 to 6-hour (n = 40) versus 2-hour (n = 20) oxaliplatin infusion duration. Eligibility includes confirmed GI malignancy, ECOG 0–2, and adequate organ function. Key exclusion criteria are baseline grade ≥2 neuropathy and prior hypersensitivity/intolerance to oxaliplatin. The primary endpoint is the between-arm difference in EORTC Chemotherapy Induced Peipheral Neuropathy (CIPN)20 sensory scores from baseline to cycle 4. Secondary endpoints include comparison of pharmacokinetic parameters, particularly Cmax of platinum ultrafiltrate. With an assumed 10-point between-group difference in CIPN20 sensory score at cycle 4, 60 patients provide 80% power (two-sided α = 0.05). The study is open and accruing (NCT03800693). Clinical trial information: NCT03800693 .

Synergistic effects of CaCO₃ nanoparticles on the properties of clay-based hybrid nanocomposites chitosan/montmorillonite: Photocatalytic, electrical, and dielectric performance

Beyond the Cr ³⁺ Paradigm: Bi ²⁺ ‐Enabled Robust and Self‐Recoverable Near‐Infrared Mechanoluminescence for Mechanically Adaptive Optoelectronics

ABSTRACT Near‐infrared (NIR) mechanoluminescence (ML) materials that directly convert mechanical stimuli into optical signals are highly desirable for mechanically adaptive optoelectronics. To date, however, this field has been overwhelmingly dominated by Cr 3+ ‐based phosphors, whose emission relies on piezoelectric‐field‐driven excitation. While effective in rigid inorganic hosts, such systems suffer from pronounced performance degradation when transferred into flexible matrices, resulting in poor repeatability, limited cyclic stability, and restricted applicability. Here, we move beyond the Cr 3+ paradigm by introducing Bi 2+ as an alternative and fundamentally distinct activator for flexible NIR ML. By incorporating a newly developed Sr 3 (BO 3 ) 2 :Bi 2+ phosphor into a polydimethylsiloxane (PDMS) matrix, a chromium‐free NIR ML elastomer is developed, leveraging a triboelectrification‐induced interfacial charge transfer mechanism to achieve robust broadband emission peaking at 815 nm under diverse mechanical excitations. The composite demonstrates highly repeatable and cyclically stable NIR ML over 10,000 continuous stretching cycles, with an impressive initial power density (29.0 mW·m −2 ) and self‐recovery behavior (24.1%@1 min and 91.5%@24 h). Notably, its initial ML intensity exceeds that of state‐of‐the‐art Cr 3+ ‐based counterparts (e.g., Ga 2 O 3 :Cr 3+ /PDMS) by more than 2.2‐fold, while simultaneously exhibiting markedly enhanced cyclic stability. This work presents a chromium‐free, self‐powered, and self‐recoverable NIR ML system, paving the way for mechanically adaptive optoelectronic devices.

Perceptions of supportive care among patients with stage IV or recurrent cancer in a community oncology setting.

e24078 Background: Early integration of supportive and palliative care (SC/PC) improves outcomes for patients with advanced cancer; however, patient stigma, misconceptions, and communication gaps may delay uptake. Despite strong evidence supporting early PC, patient perceptions and utilization of SC/PC remain understudied. Prior studies suggest that patients and caregivers are receptive to supportive care services, yet less likely to engage when palliative care terminology is used. Understanding patient perspectives and patterns of SC/PC utilization may inform strategies to improve access and quality of care. Methods: From June 2025 to January 2026, we conducted a prospective, multi-site survey of adults with stage IV or recurrent malignancies at Hartford HealthCare. English- or Spanish-speaking patients completed a 10–15 minute survey assessing demographics, support and advance care planning, SC/PC utilization, and perceptions of palliative care using a validated instrument. Descriptive statistics were used for analysis. Results: A total of 163 surveys were completed. Mean age (±SD) was 67.5 ± 11.8 years, and most respondents were 1–2 years from diagnosis. Overall, 74.8% reported discussing SC/PC with a healthcare provider. Across 10 available SC services, 241 instances of utilization were reported, with markedly higher use among those who discussed SC/PC (87.6%) versus those who had not (12.4%). Among patients with provider discussion, the most utilized services were nutrition (47.4%), social work (34.8%), and home care (25.5%). In contrast, patients without SC/PC discussion commonly used physical/occupational therapy (21.2%) and nutrition (12.9%), with minimal use of palliative care (0%), integrative medicine (0%), and behavioral health (6.3%). Hospice care was explained to 33.3% of patients, and 27.9% reported understanding when it may be appropriate. Reported barriers included feeling overwhelmed by information (20.8%), limited understanding of services (18.2%), and financial stress (14.2%). Median readiness to hear about PC was 4 (0–10 scale), with 26.9% reporting 0 and 13.1% reporting 10. Conclusions: SC/PC discussions between patients and healthcare providers demonstrated higher and more varied service utilization. Although engagement in SC services increased when discussed by the oncology team, overall utilization remained suboptimal and may be limited by information overload, lack of patient understanding, persistent stigma, or financial stress. Targeted patient education regarding SC/PC, along with improved coverage, may enhance readiness and utilization of SC/PC services.

Is adjuvant chemotherapy associated with improved survival for patients with a complete pathological response to neoadjuvant chemotherapy for pancreatic adenocarcinoma?: A population-based analysis.

e16453 Background: The treatment for localized pancreatic adenocarcinoma often involves neoadjuvant chemotherapy and surgical resection followed by adjuvant chemotherapy. In the case of a complete pathologic response (pCR) with node-negative disease, it is unclear if adjuvant chemotherapy improves survival. This study aims to identify whether adjuvant therapy for patients with a pCR is associated with improved survival. Methods: This study utilizes the 2004-2022 National Cancer Database participant user file on pancreatic cancers. Patients were selected for a histological diagnosis of pancreatic adenocarcinoma, age ≥18 years, treated with neoadjuvant chemotherapy and underwent either a pancreaticoduodenectomy or distal pancreatectomy. Patients were excluded if they had a Charlson-Dayo score greater than 2, AJCC staging greater than ypT0, treated with either hormone or immune therapy, or died within 90 days of surgery. IBM SPSS was used to create two cohorts, those that received adjuvant therapy and those that did not. SPSS was used for Kaplan Meier analysis to compare the survival and follow-up between the two groups and log-rank analysis to illicit significance. Results: Out of 9748 patients, a total of 508 patients had a pCR (5.5%). In the adjuvant cohort (n = 109), the 5-year survival was 58.6% and the 10-year survival was 34.3%. In the no therapy cohort (n = 399), the 5-year survival was 58.1% and the 10-year survival was 38.3%. The median survival of the adjuvant cohort was 94.720 ± 23.241 months, and the median survival of the no therapy cohort was 78.260 ± 8.856 months. Log rank analysis yielded a p-value of .865. The median follow-up of the adjuvant therapy cohort was 61.630 ± 5.907 months, and the median survival of the no therapy cohort was 54.800 ± 1.826 months. Log rank analysis yielded a p-value of .298. Conclusions: For patients who had a pCR after neoadjuvant chemotherapy for pancreatic adenocarcinoma, adjuvant chemotherapy is not associated with an improvement in survival.

Definitive non-surgical management of locally advanced esophageal adenocarcinoma: A systematic review and meta-analysis.

e16143 Background: For patients with Locally Advanced Esophageal Adenocarcinoma (LAEA), esophagectomy with perioperative chemo-radiation is the preferred treatment. Esophagectomy, however, is associated with substantial morbidity, with postoperative complication rates approaching 60%. Organ-preserving, esophagectomy-free approaches may reduce treatment-related morbidity but carry a risk of locoregional recurrence. This meta-analysis evaluates the efficacy and safety of esophagectomy-free treatment approaches for LAEA. Methods: A comprehensive PRISMA-guided systematic literature search was conducted across PubMed, Cochrane, and Embase, yielding 8 clinical trials evaluating non-surgical treatment strategies in patients with LAEA. Data were extracted, and outcomes were pooled using inverse-variance meta-analysis with logit transformation under random-effects models, with heterogeneity quantified by I² in the R statistical package version 4.4.0. Results: A total of 945 patients were included, of whom 76% were male and 24% female, with a median age of 65.5 (58-72) years. Patients had non-metastatic locally advanced stage II–III disease and good baseline functional status (ECOG/WHO/Zubrod 0–1). Patients were treated with cetuximab plus chemoradiation (CRT) = 413, CRT with weekly carboplatin+paclitaxel = 129, induction chemotherapy (IC) + CRT = 51, pembrolizumab plus CRT = 40, XELOX followed by CRT = 19, IC + radiotherapy (RT) = 17, definitive CRT = 12, definitive RT = 8, and celecoxib plus CRT = 7. The median follow-up period was 22.3 (16-46) months. The pooled complete response rate was 41% (95% CI, 24–60%; I² = 93.3%), and the pooled overall response rate was 42% (95% CI, 32–53%). Pooled 2-and 3-year overall survival rates were 45% (95% CI, 40-49%) and 36% (95% CI, 29-45%; I² = 75.7%), respectively. Median OS was 19.3 (16-47) months. The pooled 2-year PFS was 32% (95% CI, 7-74%; I² = 82.2%), with a median PFS of 11.8 (8.8-24.1) months. Pooled rates of locoregional and distant progression were 41% (95% CI, 33–49%; I² = 65.7%) and 36% (95% CI, 25–48%; I² = 79.1%), respectively. Treatment-related adverse events were variably reported; common grade ≥3 toxicities included hematologic events, pneumonitis, esophagitis, dysphagia, diarrhea, fatigue, and dehydration. The pooled incidence of grade 3–4 neutropenia was 12% (95% CI, 6–22%; I² = 72.5%). Conclusions: Esophagectomy-free treatment strategies demonstrate meaningful clinical activity with acceptable toxicity in selected patients with LAEA. These approaches may offer a viable organ-preserving alternative to standard therapy. Prospective randomized trials are needed to better define patient selection, optimize surveillance strategies, and compare long-term outcomes with surgical approaches.

NICE-TIL (NKG2D-CAR/mbIL15/tEGFR-transduced TILs) and antitumor activity via dual TCR-HLA and NKG2D-NKG2DL killing pathways.

e21506 Background: CAR-T therapy’s efficacy in solid tumors is limited by T-cell dysfunction, heterogeneous antigen expression, and safety concerns. NICE-TIL (NKG2D-CAR-engineered TILs) integrates: 1) TILs inherent tumor-homing capacity and diversified TCR to tumor cell antigens; 2) NKG2D’s targeting of NKG2D ligands (NKG2DLs) broadly expressed on tumor cells and downstream signal activation; 3) membrane-bound IL-15 (mbIL15) for TME-resilient survival; 4) truncated EGFR (tEGFR) as a safety switch 5). In this study, we present data of NICE-TIL in vitro tumor killing activities and first in human clinical results. Methods: A: in vitro study: NICE-TIL and conventional TILs (CTR-TIL) were tested for 1) tumor cell killing activities to: different cancer cell lines and patient-derived organoids; 2) cell Proliferation activities after cancer cell killing in CD3 + /CD4 + /CD8 + subsets; 3) cancer cell killing mechanisms via blocking antibody (HLAI/NKG2D) study. 4) Safety assessment via tEGFR-mediated depletion. B: first-in-human clinical report. Results: 1) NICE-TIL exhibited superior cytotoxicity vs. CTR-TIL: Cell lines: >60% lysis (3rd sequential co-culture round) vs. <30% of CTR-TIL (melanoma/lung/cervical cell lines).Organoids: 80% (10:1 E:T) vs. 60% (CTR-TIL) for lung organoids; 2) Proliferation (Ki-67 + %) in NICE-TIL subsets was markedly enhanced by tumor co-culture (CD3 + : 60% vs. 20% without tumor; CD4 + : 65% vs. 20%; CD8 + : 60% vs. 20%). 3) Blocking antibody assays confirmed dual TCR-HLA and NKG2D-NKG2DL killing pathways (dual blockade reduced cytotoxicity to 45% vs. 90% in control). 4) Safety assessments: tEGFR-mediated depletion eliminated >95% of NICE-TIL in vitro. NICE-TIL demonstrates potent, sustained activity across different solid tumor models (cell lines + organoids) via dual killing pathways. Its tEGFR safety switch and lack of off-target toxicity address key clinical concerns. This engineered TIL platform leverages NKG2D’s broad targeting, mbIL15’s pro-survival effects, and TILs’ tumor-homing capacity, supporting its translation for melanoma, lung, cervical, and pancreatic cancer. The NICE-TIL therapy has entered the clinical trial phase. The first subject was a patient with advanced melanoma who had developed resistance to anti-PD-1, anti-VEGFR and chemotherapy, and achieved a confirmed partial response (PR), with 63% reduction in tumor burden. Conclusions: In summary, both in vitro data and clinical result suggest NICE-TIL may be an effective treatment for late-stage solid tumors.

Prospective prediction of alisertib-induced Aurora-A kinase resistance mutations in patients with metastatic ER+ breast cancer using an innovative artificial intelligence approach.

3022 Background: Pharmacological blockade of Aurora-A mitotic kinase (AURKA) oncogenic pathway represents a promising therapeutic strategy to inhibit breast cancer plasticity and progression. However, treatment with the selective AURKA inhibitor alisertib is followed by acquired on-target mutations in a subset of patients that are linked to alisertib resistance and tumor progression. Current resistance mechanisms are typically identified retrospectively after therapeutic failure. We evaluated whether Dynaptive, a structure- and evolution-aware artificial intelligence (AI) platform, could prospectively predict treatment-induced AURKA resistance mutations prior to experimental observation. Methods: We identified acquired on-target mutations in patients from the alisertib Phase-II clinical trial. These mutations were cross-matched against Dynaptive predictions that had been generated independently of any experimental resistance data. Dynaptive integrates protein conformational ensembles and evolutionary sequence constraints to predict mutational hotspots and assess their functional and drug-binding impact. Identified mutations were evaluated against large-scale population datasets to determine natural occurrence. Results: Several AURKA mutations observed in patients that showed resistance to alisertib were prospectively predicted by Dynaptive despite being entirely absent from prior population datasets. These mutations were not observed previously in either healthy or cancer cohorts described in the public domain. Significantly, the most frequently occurring AURKA resistance mutation occurred at a residue ranked second among predicted mutational hotspots and were predicted to reduce alisertib binding affinity by ≥10-fold. Across all variants, absence from population datasets and emergence only following treatment support a treatment-induced resistance mechanism. Conclusions: A structure-aware AI platform prospectively identified clinically relevant, treatment-induced AURKA resistance mutations prior to experimental detection. These findings demonstrate the feasibility of predictive modeling to anticipate resistance mechanisms and support the development of next-generation AURKA inhibitors optimized to overcome mutations-driven resistance and to improve therapeutic efficacy, safety and durability.

Impact of engineered probiotic expressing FomA-targeted antimicrobial peptides on intestinal tumorigenesis in a mouse model of <i>Fusobacterium nucleatum</i> –associated colorectal cancer.

e15157 Background: Fusobacterium nucleatum ( Fn ) is one of the most enriched bacteria in colorectal adenomas and adenocarcinomas, where it promotes tumor progression, suppresses anti-tumor immunity, and contributes to chemoresistance. Current antibiotic approaches to eliminate Fn cause collateral damage to beneficial gut microbiota and diminish therapeutic efficacy. To eliminate Fn directly, we engineered Lactococcus lactis MG1363 to express guided antimicrobial peptides (gAMPs) comprising Cathelin-related peptide CPSC5 linked to a Statherin-derived guide peptide (YQPVPE) that specifically binds FomA, an outer membrane porin and key virulence factor of Fn . Building on our previous in vitro findings, which demonstrated that gAMPs selectively inhibited F n growth and biofilm formation, this study sought to address efficacy in vivo. Methods: We tested our engineered probiotic in the Apc^Min/+^ mouse model. We pre-treated 32 male and female mice with streptomycin (2 mg/mL/d x7d) followed by DSS (1.5% x7d) and colonized with the 7_1 Fn strain (1 × 10 9 cells/ 200 µl x1d), which has shown to promote CRC in animal models. Results: Preliminary results demonstrate striking tumor suppression across all L. lactis treatment groups compared to untreated controls. Control mice exhibited 100% tumor incidence (4/4 mice) with 32 total tumors. Treatment with empty vector L. lactis alone significantly reduced tumor incidence to 33% (2/6 mice) with only 4 total tumors, consistent with prior reports of L. lactis immune stimulatory properties, while unguided AMP- L. lactis showed 83% tumor incidence (5/6 mice) with 22 total tumors. Most strikingly, gAMP- L. lactis treatment achieved the greatest tumor suppression, reducing incidence to just 17% (1/6 mice; p&lt;0.05 vs. control) with only a single tumor observed across all treated animals. These data demonstrate that FomA-targeted gAMP delivery provides superior efficacy over both empty vector probiotic and unguided AMP approaches. Conclusions: Ongoing studies are examining the effects of gAMP-probiotic treatment further, on gut microbiome composition and function, anti-tumor immune responses, expression of oncogenic signaling genes, and Fn eradication. Overall, these data suggest the therapeutic potential of engineered probiotics secreting pathogen-targeted antimicrobial peptides as a precision approach to prevent and treat cancer.

Minimal residual disease ctDNA testing in metastatic breast cancer patients with no radiological evidence of disease: Potentialclinical utility from real-worlddata.

e13075 Background: Liquid biopsy is increasingly recognized as a valuable tool for disease monitoring in metastatic breast cancer (MBC), with growing evidence supporting its role in treatment response assessment and minimal residual disease (MRD) detection across disease settings. This multi-institutional retrospective analysis evaluated the real-world use of circulating tumor DNA (ctDNA)–based MRD testing in patients(pts) with MBC who achieved radiological no evidence of disease (NED). Methods: We retrospectively analyzed pts with MBC treated at four institutions within the Precision Medicine Action for Cancer (PMAC) consortium who achieved radiological and/or metabolic complete response (CR) PET or CT imaging at any point during their MBC course. ctDNA-based MRD assessment was performed as part of routine clinical practice between September 2022 and December 2025 using plasma-only (Guardant Reveal, Guardant Health) or tissue-informed (Signatera) assays. Data were collected through retrospective review of electronic medical records under appropriate IRB approval. Analyses were descriptive. Results: 43 pts with MBC who achieved radiological NED and underwent MRD testing were identified, accounting for 225 ctDNA tests (median 5 tests per patient; IQR 1–8). Molecular subtypes included 36% luminal disease,24% triple-negative, and 40% HER2-positive disease. Median follow-up (fup) from primary breast cancer diagnosis was 63.2 months (IQR, 34.1–109.4). Longitudinal testing (≥2 ctDNA tests) was performed in 22 patients (51%). At first MRD assessment, 35 pts (81.4%) were MRD-negative: 27 had already achieved radiological CR, while in 8 patients MRD negativity preceded radiological NED by a median of 2.7 months (range 0.4-13.4). 5 pts had a positive MRD result at first evaluation concordant with radiological evidence of disease, and 3 had a positive first MRD test despite the absence of detectable disease. Among pts with at least one post-response MRD assessment, 26 pts (70.3%) showed persistently negative MRD results during longitudinal monitoring; among them, imaging was deferred in 7 cases and treatment de-escalation was pursued in 13 cases after at least one negative MRD test. Overall, 14 pts (32.6%) had at least one MRD-positive result during response fup; in six cases, longitudinal ctDNA monitoring anticipated radiological disease progression by a median of 1.05 months (range 0.5-3.7). At last fup, 20 pts remained radiologically disease-free with persistently negative ctDNA while in CR (median fup from first negative ctDNA test of 15.3 months). Conclusions: ctDNA MRD negativity was largely concordant with radiological or metabolic CR. This real-world cohort supports further prospective investigation to inform clinical management in patients with MBC achieving radiological NED.

Beyond cellular locality: Vision transformer–based modeling for diagnosis and classification of blood cancers from peripheral blood smears.

e18562 Background: Accurate morphologic assessment of peripheral blood smears remains central to hematologic malignancy diagnosis but is time-intensive and subject to interobserver variability, particularly in early or morphologically ambiguous disease. Convolutional neural networks (CNNs) have demonstrated utility in hematologic image analysis; however, their reliance on localized receptive fields may limit modeling of global cellular context and spatial relationships. Vision Transformers (ViTs) leverage self-attention mechanisms to enable long-range contextual reasoning across entire images. We evaluated the performance of a transformer-based architecture for automated classification of hematologic malignancies from peripheral blood smear images. Methods: This retrospective diagnostic modeling study utilized publicly available, de-identified peripheral blood smear image datasets derived from ALL-IDB–based repositories, including acute lymphoblastic leukemia, additional leukemic subtypes, and normal hematologic controls annotated by expert hematopathologists. Images were standardized, augmented, and divided into training and validation cohorts using stratified sampling. A pretrained Vision Transformer B/16 architecture was fine-tuned for multi-class classification. Input images (224×224 pixels) were partitioned into 16×16 patches and processed through 12 transformer encoder layers with multi-head self-attention. Model performance was assessed using accuracy, sensitivity, specificity, F1 score, and area under the receiver operating characteristic curve (AUROC). Performance was compared with published CNN benchmark results on similar datasets. Results: The Vision Transformer demonstrated high diagnostic performance across hematologic categories, achieving overall classification accuracy exceeding 99% with AUROC values greater than 0.99. Attention-based global modeling improved discrimination of leukemic blast populations with overlapping cytomorphologic features and reduced misclassification in visually heterogeneous samples. Performance remained stable across variable staining conditions and cell population distributions and was comparable to or exceeded reported CNN-based benchmarks on ALL-IDB–derived datasets. Conclusions: Transformer-based global attention modeling enables accurate and robust classification of hematologic malignancies by capturing spatial and population-level morphologic context beyond conventional localized feature extraction. These findings support the potential role of Vision Transformer architectures as AI-assisted diagnostic decision-support tools to augment hematopathology workflows. Prospective validation using real-world clinical datasets is warranted to assess generalizability and clinical integration.