Browse Articles

Real-world use of artificial intelligence for automated prognostication and systemic therapeutic decision making in localized RCC.

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

Efficacy of tocilizumab in resolving chimeric antigen receptor T-cell (CAR T)–induced cytokine release syndrome (CRS): A pooled clinical trial (CT) analysis of patients (pts) with B-cell lymphoma (BCL).

7027 Background: CAR T therapy offers significant efficacy in the treatment of relapsed/refractory hematologic malignancies but is also associated with developing immune-mediated toxicity, such as CRS. Tocilizumab, an interleukin-6 receptor antagonist, is the only FDA-approved treatment for CAR T-induced CRS, but its efficacy in large pt populations remains limited. This study aims to evaluate the efficacy of tocilizumab by assessing complete response (CR) rates and median time-to-CR in pts with BCL experiencing CAR T-induced CRS, using pooled clinical trial data. Methods: We pooled aggregated, anonymized data made available through Medidata Clinical Cloud across <10 multinational CTs, comprising 2,304 patients. Inclusion criteria required a BCL diagnosis, who developed CAR T-induced CRS and were treated with tocilizumab. The index date was defined as the initiation of tocilizumab. CR was defined as CRS resolution within 14 days of the index date without meeting any failure to respond criteria. Non-response was defined as requiring>2 doses of tocilizumab, rescue therapy with siltuximab, CRS grade escalation after 2 tocilizumab doses, or CRS recurrence after 2 tocilizumab doses. The CRS events were graded using Lee's Criteria 2014. CR rate was described as a proportion, and time-to-CR was estimated using the Kaplan-Meier method. Results: Of the 2,304 pts included in the database, 680 met the inclusion criteria (37% aged ≥ 65 yrs; 63% male; 80% white; 47% Diffuse Large BCL, 14% Follicular Lymphoma & 13% Mantle Cell Lymphoma; 49% Eastern Cooperative Oncology Group (ECOG) performance score=0 & 49% ECOG=1; 48% Ann Arbor Stage IV; 27% had bulky disease; 65% without bridging therapy; 73% received CD28 costimulatory domain-targeted CAR T-infusion, 27% received 41bb costimulatory domain-targeted CAR T-infusion; 72% with grade 1 CRS at the time of onset). The median time from CAR-T infusion to CRS onset was 2 days (IQR: 1 - 4), and the median time from CRS onset to the first dose of tocilizumab was 2 days (IQR: 1 - 3). The CR rate was 66% (95% CI: 62.5% - 69.6%) and the median time to CR was 4 days (95% CI: 3 - 4), with 35% achieving CR in ≤ 2 days and 80% in ≤ 7 days. Conclusions: This large-sample pt-level study of pooled CT data showed significant efficacy of tocilizumab in resolving CAR T-induced CRS among pts with BCL, with two-thirds of pts achieving CR. The rapid median time-to-CR (4 days) underscores the role of tocilizumab as a first-line intervention. However, the 34% of pts who did not meet the CR criteria suggest a need for further investigation into other treatment strategies.

CD4 T-cell reconstitution and CD4:CD8 inversion following lymphodepleting chemotherapy for tumor-infiltrating lymphocyte therapy in metastatic melanoma.

9551 Background: Lymphodepleting (LD) chemotherapy is required prior to tumor-infiltrating lymphocyte (TIL) therapy, yet the depth, kinetics, and completeness of peripheral T-cell recovery in practice remain poorly defined. Using a pooled, multi-institution cohort, we characterized longitudinal CD4 and CD8 T-cell reconstitution and the prevalence of CD4 lymphopenia following TIL therapy. Methods: We conducted a retrospective multicenter study across Mayo Clinic, Moffitt Cancer Center, and Massachusetts General Hospital of patients with metastatic melanoma treated with TIL (N = 41, median follow-up 7.6 months). Absolute CD4 and CD8 T-cell counts and CD4:CD8 ratios were assessed at baseline (pre-LD), week 4, month 3, and month 6 post-TIL. Clinically significant CD4 lymphopenia was defined as CD4 < 200 cells/µL. All patients received Pneumocystis jirovecii and antiviral prophylaxis, with antibacterial and antifungal prophylaxis administered per institutional protocols. Post-TIL infections were recorded. Paired changes from baseline were evaluated using Wilcoxon signed-rank testing. Results: Median age was 65 years (range 35-79), and 68% were male. At baseline, 16 patients (39%) had CD4 counts below the institutional lower limit of normal prior to LD. Eighteen patients (44%) received bridging therapy prior to TIL and had lower baseline CD4 and CD8 counts compared with those who did not ( p < 0.05 for both); however, post-TIL CD4 and CD8 recovery trajectories did not differ between groups. CD4 counts declined sharply following LD and demonstrated incomplete and heterogeneous recovery. Median CD4 decreased from 490 cells/µL at baseline to 250 at week 4, 290 at month 3, and 220 at month 6, remaining significantly below baseline at all post-TIL timepoints ( p < 0.05). Clinically significant CD4 lymphopenia was observed in 45% (10/22) of evaluable patients at month 3 and in 50% (44%, 7/16) at month 6. In contrast, CD8 counts increased early after TIL (median 240 to 590 cells/µL at week 4; p = 0.004) and remained higher than baseline at month 6. The CD4:CD8 ratio inverted from a median of 2.02 at baseline to 0.46 at week 4 and remained below 1.0 through month 6, reflecting sustained CD4 suppression despite CD8 recovery. Post-TIL infections occurred despite widespread prophylaxis, including bacterial (13%), viral (10%), and fungal (4.9%) infections. Conclusions: In a multi-institution cohort, LD chemotherapy preceding TIL is associated with profound and prolonged CD4 T-cell suppression, persistent CD4:CD8 inversion, and a high prevalence of CD4 < 200 cells/µL through 6 months. These findings identify delayed immune reconstitution as a common and clinically relevant consequence of TIL therapy and support prospective optimization of antimicrobial prophylaxis duration, vaccination timing, and immune monitoring strategies.

Transportation insecurity and RBC transfusion burden in multiple myeloma and related plasma cell disorders in New Mexico: A CRIIS statewide analysis (2013-2019).

1532 Background: Multiple myeloma (MM) and related plasma cell disorders are commonly complicated by anemia requiring red blood cell (RBC) transfusion. In geographically large and predominantly rural states such as New Mexico, transportation insecurity may contribute to fragmented outpatient access and downstream supportive care burden. Using statewide oncology informatics data, we characterized RBC transfusion utilization patterns and their association with county-level vehicle access. Methods: Using the Cancer Research Integrated Informatics System (CRIIS), a novel statewide oncology informatics platform, we identified patients with MM and related plasma cell disorders (including Waldenström macroglobulinemia and plasma cell leukemia) receiving care from 2013-2019. We performed a repeated cross-sectional county-level analysis, using county-year as the analytic unit. Transportation insecurity was defined as the percent of households without vehicle access and summarized using the median county value across 2013-2019 due to minimal year-to-year variability. Counties were categorized as higher versus lower transportation insecurity using a median split. The primary outcome was receipt of ≥1 RBC transfusion within a calendar year, summarized at the county level. Results: Among 2,436 patients statewide, 614 (25.2%) received ≥1 RBC transfusion. Annual transfusion utilization increased from 15.0% in 2013 and stabilized at ~25-30% thereafter. Median county transportation insecurity ranged from 0% to 12.8%. Across multiple years, counties with higher transportation insecurity accounted for a disproportionate share of statewide transfusion burden relative to patient volume (~1.3-1.5-fold higher absolute transfusion burden). Transfusion burden was concentrated in a subset of counties, with Bernalillo contributing the largest share, while several smaller counties demonstrated high transfusion rates despite low total patient volume. Transfusion delivery remained concentrated in access-rich counties, consistent with regional centralization of oncology supportive care. Conclusions: In this statewide real-world analysis, RBC transfusion burden in MM and related plasma cell disorders clustered disproportionately among residents of transportation-insecure counties. Transportation insecurity may serve as a pragmatic structural marker to identify regions bearing elevated supportive care burden and supports collaborative evaluation of transportation-focused care delivery strategies to improve equity across rural and underserved communities. Statewide cohort size and RBC transfusion utilization by year. Year Total Patients Transfused Patients Transfusion Rate (%) 2013 240 36 15.0 2014 238 49 20.6 2015 301 67 22.3 2016 373 109 29.2 2017 411 118 28.7 2018 427 107 25.1 2019 446 128 28.7 Total 2,436 614 25.2

First-in-human clinical evaluation of ST-01156, an optimized and selective degrader of RNA-binding motif 39 (RBM39): A phase 1 study in advanced solid malignancies with a focus on RBM39-dependent cancers.

TPS3164 Background: ST-01156 is an orally administered agent developed to degrade RBM39, a protein target that is principally involved with splicing RNA into messenger RNA and is commonly upregulated in cancer. ST-01156 has entered phase 1 clinical trials for the treatment of several cancer types supported by its novel mechanism of action and data from preclinical testing. The lead optimization process for ST-01156 maximized degradation selectivity, anticancer potency, metabolic stability (including deuteration), and other drug-like properties. The ability of ST-01156 to achieve functionally meaningful and selective effects is exemplified by complete degradation of RBM39 within hours following treatment in vitro, accompanied by downstream mis-splicing of pathogenic transcripts, most notably the EWSR1-FLl1 fusion transcript driving Ewing sarcoma (ES). Adding to this effect, ST-01156 depleted proteins involved in DNA damage repair (DDR), supporting the potential for synthetic lethality with cancer cells that depend on proficient DDR. RBM39 is also involved in the splicing of KRAS isoforms thought to be essential for the proliferation and survival of oncogenic KRAS cancer stem cells, providing an anticancer mechanism distinct from the mechanisms of current KRAS inhibitory agents. In advanced hepatocellular carcinoma (aHCC), RBM39 is substantially overexpressed and likely impedes essential arginine sensor activities. In addition to demonstrating robust activity in these malignancies, ST-01156 is highly active in patient-derived xenograft models of neuroblastoma, biliary tract carcinoma (BTC), and endometrial carcinoma (EC). In animal toxicology studies, hematologic and gastrointestinal tissues were most susceptible to RBM39 inhibition. Methods: In the first-in-human study, ST-01156 doses will be adaptively escalated based on the rate and severity of adverse events in trial participants with advanced solid malignancies. ST-01156 will be administered orally for 5 consecutive days every 7 days with an option of adapting a continuous dosing schedule, if tolerable. Backfilling of safe dose levels with participants with high priority malignancies will be allowed. In addition to traditional safety and tolerability endpoints, dose escalation and Phase 2 dose derivation will be based on pharmacokinetic (PK) endpoints and real-time RBM39 targeting measurements in peripheral mononuclear cells. The optimal dose of ST-01156 derived from safety tolerability, PK and pharmacodynamic data will be evaluated in trial participants in the following expansion cohorts: A, ES; B, aHCC; C, KRAS -mutant cancers; and D, an adaptive cohort of cancers projected to be susceptible to RBM39 inhibition including BTC, EC, and cancers with DDR aberrations. Clinical trial information: NCT07197554 .

Failure-to-rescue after acute cardiac events in cancer versus non-cancer hospitalizations.

e23241 Background: Acute cardiac complications are major drivers of inpatient mortality, yet outcomes after theseevents may differ in patients with cancer due to reduced physiologic reserve, treatment-relatedcardiotoxicity, and competing critical illness. Failure-to-rescue (FTR), defined as death after acomplication, provides a systems-level framework to quantify vulnerability and gaps in careescalation. We evaluated whether mortality after acute cardiac complications is higher inhospitalizations among patients with cancer compared with non-cancer hospitalizations. Methods: We performed a serial cross-sectional analysis of the 2018–2022 National Inpatient Sampleusing discharge-level survey weighting. Adult hospitalizations were classified as cancer(principal malignancy) or non-cancer. Acute cardiac complications were identified usingICD-10-CM codes for acute myocardial infarction (AMI; I21–I22), arrhythmia (I47–I49), andshock. FTR was defined as in-hospital mortality among hospitalizations with each complication.Survey-weighted multivariable logistic regression estimated the association between cancerstatus and FTR within each event cohort, adjusting for demographics, payer, ZIP-code incomequartile, elective status, year, and hospital characteristics. Results: Acute cardiac complications were common (AMI 4.43%; arrhythmia 19.01%). In unadjustedanalyses, FTR after AMI was higher in cancer versus non-cancer hospitalizations (19.1% vs9.0%), as was FTR after arrhythmia (8.3% vs 6.0%). In the composite cohort of any acutecardiac complication, hospitalizations among patients with cancer demonstrated higher FTRmortality (10.0% vs 6.1%).After adjustment, cancer status remained associated with higher FTR after AMI (adjusted oddsratio [aOR] 2.26, 95% CI 2.13–2.39) and arrhythmia (aOR 1.67, 95% CI 1.63–1.71). In thecomposite cohort, cancer was associated with nearly doubled odds of death (aOR 2.00, 95% CI1.95–2.04). Cancer status could not be independently estimated in shock-only admissions because of collinearity. Cancer significantly modified the association between cardiaccomplications and mortality (interaction OR 1.16, 95% CI 1.12–1.20). Conclusions: Hospitalizations among patients with cancer demonstrate substantially higher failure-to-rescueafter acute cardiac complications, particularly following myocardial infarction and arrhythmia.These findings highlight FTR as a practical inpatient quality metric and highlight the need forearlier recognition, aggressive escalation, and standardized rescue pathways for cardiacdecompensation in oncology populations.

Development and application research of the intervention program for medication compliance of endocrine therapy in breast cancer patients based on the ICAMO model.

e24099 Background: Hormone receptor-positive breast cancer is the most common cancer among women globally. Adjuvant endocrine therapy (AET) is a key treatment that reduces the risk of recurrence and improves survival rates. However, medication adherence remains a significant challenge, influenced by multiple factors. Existing research often focuses on measuring the effectiveness of interventions but tends to overlook the contextual and mechanistic aspects that influence their success. Objective: This study aims to explore how to optimize adherence interventions for breast cancer patients undergoing AET, analyze the mechanisms and contextual factors involved, and verify the effectiveness of these interventions through clinical practice. Methods: The study uses Realist Synthesis (RS) methodology, analyzing the ICAMO model (Intervention-Context-Actors-Mechanism-Outcomes) to understand the mechanisms behind adherence interventions. Data was integrated from systematic literature reviews, expert interviews, and stakeholder feedback to identify the key mechanisms driving successful outcomes and evaluate their effectiveness in promoting AET adherence. The evidence was then translated into clinical practice through an evidence-based clinical translation model to validate the intervention’s effectiveness. Results: The findings from the RS revealed that key mechanisms for improving adherence include enhancing patient beliefs about treatment efficacy, improving symptom management, providing behavioral cues, and offering emotional and social support. Contextual interventions tailored to individual patient needs and specific circumstances were found to be more effective than generalized approach. Particularly, interventions combining psychological support and symptom management had a greater impact on long-term adherence. Clinical practice verification showed that personalized interventions and regular follow-ups significantly improved adherence. The interventions included regular medication reminders, education on side effects, psychological support, and enhancing patient self-efficacy. Continuous involvement of healthcare providers and emotional support played a critical role in improving adherence. Conclusions: The application of the ICAMO model integrates various intervention strategies into a comprehensive framework, providing a holistic solution for improving AET adherence in breast cancer patients. By focusing on the mechanisms and contextual factors influencing adherence, the implementation of personalized and multi-layered intervention strategies plays a crucial role in enhancing medication adherence, ultimately improving patient outcomes in breast cancer care. This study provides important insights for optimizing clinical practice and policy recommendations. Clinical trial information: IRB#2021-07-05.

Frequent monitoring of NSCLC immunotherapy using an mDETECT liquid biopsy to identify therapy response and progression compared to CT scans.

e14532 Background: Response to immunotherapy in metastatic non-small cell lung cancer (NSCLC) is heterogeneous, and early identification of non-responders remains a clinical challenge. The methylation DETEction of Circulating Tumour DNA (mDETECT) assays are targeted DNA methylation-based Next Generation Sequencing liquid biopsies. We have developed a version of our mDETECT assay that sensitively and quantitatively monitors NSCLC. The mDETECT NSCLC assay has a 95% sensitivity at 95% specificity with an AUC of 0.95. With only 2 million reads required per sample, multiple samples can be sequenced at the same time allowing for more frequent testing. Methods: We conducted a pilot observational study to frequently measure tumour dynamics in patients undergoing first-line pembrolizumab monotherapy for metastatic NSCLC. 19 participants were recruited prior to the initiation of immunotherapy with blood being collected pretreatment then weekly or biweekly after treatment initiation with some patients being followed for up to 2.3 years. In total we collected 226 samples (median 11.8 timepoints per patient). Radiological assessment was performed approximately every three months as per the standard of care. Results: The primary aim of the study was to determine if overall survival (OS) could be predicted with the mDETECT assay within the first 6 weeks of treatment. Patients were divided by short ( < 1 year), medium (1-4 years), and long term ( > 4 years) OS. Patients with short OS showed constant or increasing mDETECT levels and never dropped below 80% of their pre-treatment mDETECT level. Patients with long OS showed an immediate decrease within the first 6 weeks of treatment and generally reached undetectable levels. Patients with medium OS showed a slower decline and a higher steady state level than the long OS patients. Continued monitoring revealed progression in some initially responding patients, with increasing mDETECT levels detected 4-6 months in advance of radiological progression. Eight patients had CT imaging available for analysis. Five of eight patients had concordance between ctDNA kinetics and radiological response to immunotherapy. Conclusions: These patterns, while derived from a small pilot cohort, suggest the mDETECT assay can determine within weeks of treatment initiation if a patient is responding to immunotherapy. Frequent assessment of tumour burden by a liquid biopsy such as mDETECT offers the opportunity to modify treatments earlier than radiology alone and to do so on an ongoing basis. These findings support the feasibility and potential clinical utility of integrating frequent methylation-based ctDNA monitoring into immunotherapy management workflows and justify further prospective validation in a larger study.

Predictive circulating small extracellular vesicle miRNAs for immunochemotherapy response in ES-SCLC.

e20146 Background: Extensive-stage small cell lung cancer (ES-SCLC) is a highly aggressive malignancy with limited treatment options. While immunochemotherapy constitutes standard first-line therapy, predictive biomarkers for treatment response remain undefined, impeding personalized therapeutic strategies. Methods: We investigated circulating small extracellular vesicle (sEV)-derived microRNAs (miRNAs) as non-invasive predictive biomarkers in treatment-naïve ES-SCLC patients receiving immunochemotherapy. Plasma samples from a training cohort (n = 33) and an independent prospective cohort (n = 5) were collected pre-treatment. sEVs were isolated, characterized, and subjected to small RNA sequencing to quantify miRNA expression. Results: Small RNA sequencing revealed 23 differentially expressed sEV miRNAs between responders (n = 19) and non-responders (n = 14). Machine learning refined these candidates into a predictive model. Recursive feature elimination (RFE) yielded an 11-sEV-miRNA signature. The top-performing model (Extra Trees Gini) incorporated 5 sEV miRNAs and 1 clinical feature, demonstrating high predictive accuracy in the training set (AUC = 0.855, sensitivity = 95%, specificity = 80%). Preliminary validation in the prospective cohort achieved 80% accuracy (4/5 correct classifications). Conclusions: We established a novel sEV-miRNA biomarker panel that robustly predicts immunochemotherapy response in ES-SCLC. High discriminatory performance and initial prospective validation underscore its clinical utility for guiding treatment decisions and optimizing outcomes by avoiding ineffective therapy in non-responders.

NGS-based <i>DPYD</i> diplotype analysis beyond conventional targeted variant testing in an ethnically diverse cancer patient cohort.

e15131 Background: Germline variants in the highly polymorphic DPYD gene are the main cause of dihydropyrimidine dehydrogenase (DPD) deficiency and are strongly associated with fluoropyrimidine-related toxicity. PCR-based assays test limited predefined variants and may miss clinically relevant alleles, particularly in ethnically diverse populations. Next-generation sequencing (NGS) allows broader and comprehensive assessment of DPYD variation. Methods: This retrospective study included 4,380 diagnosed cancer patients of Caucasian and non-Caucasian ancestry undergoing routine DPYD testing using NGS-based profiling with the Shield assay at Datar Cancer Genetics. Identified variants were translated into diplotypes and assigned activity scores according to CPIC and DPWG guidelines, classifying patients as normal, intermediate, or poor metabolizers. Results: Clinically relevant DPYD variants associated with reduced or absent DPD activity were identified in 4.8% (211/4380) of patients, comprising 4.7% (n = 209) intermediate and 0.1% (n = 2) poor metabolizers. Reduced-function and no-function alleles detected in the cohort included *2A, HapB3, and c.557A&gt;G, either in heterozygous or compound diplotype configurations. Intermediate metabolizer status was driven by diplotypes such as *9A/HapB3 and c.557A&gt;G–containing combinations, reflecting the impact of a single reduced-function allele when paired with an otherwise normal allele. Poor metabolizer phenotypes were exclusively associated with biallelic or compound heterozygous diplotypes involving known no-function or severely reduced-function alleles, including *2A/*2A and HapB3/*2A. Notably, several clinically actionable diplotypes would not have been detected by assays limited to a small set of predefined variants, emphasizing the importance of comprehensive NGS-based DPYD profiling and diplotype-level interpretation for accurate phenotype assignment. Conclusions: Nearly 5% of patients were classified as intermediate or poor metabolizers, placing them at increased risk for severe fluoropyrimidine toxicity. Comprehensive NGS-based DPYD testing enables detection of both common and rare reduced-function variants, supports more accurate phenotype assignment, and provides a practical approach to safer fluoropyrimidine dosing in ethnically diverse clinical populations. Clinically actionable DPYD alleles identified by NGS. Allele Frequency Function HapB3 3.04% Reduced function allele *2A 0.91% Absent DPYD activity allele c.2846A&gt;T 0.50% Reduced function allele c.2279C&gt;T 0.23% Reduced function allele c.557A&gt;G 0.11% Reduced function allele *4 3.45% Absent DPYD activity allele *13 0.07% Absent DPYD activity allele *3 0.02% Absent DPYD activity allele *8 0.02% Absent DPYD activity allele

Retraction: SHP2 improved Late-onset fetal growth restriction via modulating ROS/BRD4/PI3K/YAP/PIGF signaling induced angiogenesis

Electroactive Metal–Organic Frameworks Enabling Unidirectional Electrochemical Capacitors and Logic Gates (MOF‐CAPode)

ABSTRACT From concept to realization, the integration of porous redox‐active metal–organic frameworks (MOFs) into asymmetric electrochemical capacitors for the assembly of electrochemical capacitor‐diodes (CAPodes) is reported. CAPodes are innovative electrochemical capacitor analogues of diodes, designed for unidirectional charge storage and logic gate applications. The novel devices deliberately utilize two distinct electroactive metal–organic frameworks with characteristic redox potentials acting as positively or negatively polarizable electrode materials, respectively. The first proof‐of‐concept devices presented here make use of the Chichibabin‐like diradicaloid formation upon oxidation of the N,N,N',N' ‐benzidinetetrabenzoate linker in DUT‐65/66 and N,N,N',N' ‐(1,4‐phenylenebis‐(azanetriyl))‐tetrabenzoate in DUT‐232/233 at high oxidation potentials as a positively polarizable electrode material paired with the highly reversible two‐step reduction in Zn(ndi) (ndi 2 − = 1,4‐bis[(3,5‐dimethyl)‐pyrazolate‐4‐yl]naphthalene‐diimide). The novel porous MOF‐CAPode achieves a remarkable figure of merit with rectification ratios (RR) up to RR I = 23 and RR II = 94% at 10 mV s −1 . The new MOF‐based CAPodes operate efficiently in “AND” and “OR” logic gates, demonstrating logic operation under varying input voltages up to 3.0 V and frequencies of up to 40 mHz.

Bio-inspired CuO nanoparticles via Crocus sativus leaf extract: Sol–gel fabrication and functional applications

Characterization of raw material in textile bone instruments from the capital of the Roman Province of Lusitania (Mérida, Spain)

Abstract Bone has long served as a versatile raw material in human societies due to its durability, workability, and availability. However, identifying the animal species used in worked bone artifacts remains challenging, as manufacturing processes and use-wear often remove diagnostic morphological features. This study applies Zooarchaeology by Mass Spectrometry (ZooMS) to investigate raw material selection in textile-related bone tools from Augusta Emerita (Mérida, Spain), the capital of Roman Lusitania, dating to the 1st − 3rd centuries CE. The archaeological context of the assemblage is particularly significant, originating from contexts consisting of a bone processing workshop for the manufacture and sale of bone objects that is one of the first installations of this type identified in the Iberian Peninsula. A total of 36 artifacts associated with textile production were analyzed where five main taxonomic groups were identified: Bos taurus , Bovidae/Cervidae, Cervus elaphus , Elephas maximus , and an unidentified avian species. These results indicate a functional selection of common domestic and wild taxa, while the presence of exotic materials points to long-distance trade networks and the potential symbolic or social significance of certain artifacts, particularly in funerary contexts.

An intricate functional relationship between NuA4 and Sfp1 regulates ribosome biogenesis in response to nutrient availability

Long-term responders to sacituzumab govitecan in metastatic triple-negative breast cancer: Real-world outcomes and predictors from a multicenter Italian cohort.

1129 Background: Sacituzumab govitecan (SG) has reshaped the treatment landscape of metastatic triple-negative breast cancer (mTNBC), demonstrating superior survival outcomes compared with standard chemotherapy in the ASCENT trial. Despite the overall poor prognosis of mTNBC, a subset of patients experiences remarkably durable responses and prolonged survival, far exceeding the median progression-free survival (PFS) observed in pivotal trial. However, the clinical characteristics and prognostic determinants of these "long-term responders" (LTRs) remain poorly defined, and real-world evidence specifically addressing this population is currently lacking. This multicenter study aims to characterize outcomes and identify potential predictors of long-term benefit from SG in a real-world setting. Methods: This multicenter observational analysis included 271 patients with mTNBC treated with SG across 18 Italian cancer centers, within a study incorporating both retrospective and prospective cohorts (NCT02284581). LTRs were defined as patients achieving a real-world PFS (rwPFS) ≥ 9 months. The primary endpoint was rwPFS, while secondary endpoints included real-world overall survival (rwOS) and objective response rate (ORR). Survival outcomes were estimated using the Kaplan-Meier method, and independent prognostic factors were identified using Cox proportional hazards models. Results: Seventy-four of 271 patients (27.3%) were identified as LTRs. Within this cohort, median rwPFS was 14 months (95% CI: 13.3–16.3) and median rwOS was 25.6 months (95% CI: 21.0–NR). The ORR was 68.9% (all partial responses; n=51), while 29.7% of patients (n=22) achieved stable disease. On multivariable analysis, previous therapy with immune checkpoint inhibitors (ICI) was independently associated with a reduced risk of progression (HR 0.47; 95% CI: 0.22–0.98; p=0.04). De novo metastatic disease also showed a trend towards a lower risk of progression (HR 0.44; 95% CI: 0.16–1.17; p=0.10). Conversely, treatment with SG in the third line compared with the second line was associated with a higher risk of progression (HR 2.04; 95% CI: 1.02–4.07; p=0.04). No significant association were observed for brain metastases (p=0.37), BRCA mutation status (p=0.33), or visceral involvement (p=0.27). Conclusions: In real-world setting, more than 25% of mTNBC patients experience a durable benefit from SG. Prior exposure to immunotherapy and earlier use of SG in the second-line setting were associated with a higher likelihood of LTR, while de novo disease showed a favourable trend. Notably, the presence of brain or visceral metastases did not preclude prolonged benefit. These findings suggest that a subset of patients with mTNBC may derive sustained clinical benefit from SG and support its use earlier in the treatment sequence. Clinical trial information: NCT02284581 .

Resolving the ambiguity between genomic rearrangements and gene fusions: An AI-augmented structural framework to define therapeutic eligibility in kinase-driven cancers.

1626 Background: A critical disconnect exists in precision oncology: clinical guidelines frequently conflate genomic "rearrangements" with functional gene "fusions", utilizing inconsistent terminology that obscures biological reality. While confirmatory RNA sequencing is recommended, it is often clinically infeasible due to tissue exhaustion or poor sample quality. Consequently, clinicians are forced to prescribe targeted therapies based on DNA-level proxies without knowing if a druggable protein actually exists. We addressed this unmet need with VeraFusionDx, an AI-augmented decision support system that moves beyond simple classification to determine therapeutic eligibility. Methods: We developed a universal framework integrating AI-driven report parsing with generative molecular modeling, trained on a massive real-world dataset (&gt;10,000 gene fusions and &gt;50,000 rearrangements) from a CAP/CLIA-certified lab. An AI normalization module standardizes heterogeneous inputs (unstructured NGS reports, gene+exon number pairs, or coordinates) into a unified format. Distinct from static database lookups, the core engine performs de novo characterization of each patient variant. It computationally reconstructs chimeric sequences and models 3D protein architecture to verify critical druggability criteria, including reading frame alignment and kinase domain integrity. Finally, an AI-literature agent cross-references validated targets with clinical evidence to define definitive therapeutic eligibility (https://verafusiondx.origimed.com). Results: Analysis using this generative engine revealed a heterogeneous landscape where genomic rearrangement proved a poor proxy for druggability, highlighting the necessity of structural validation. For NTRK genes, approximately two-thirds of detected variants were rearrangements rather than actionable fusions, with striking discordance by gene and tumor type. Notably, NTRK2 rearrangements were 4-fold more common than fusions, and FGFR1 were 90% rearrangements. Even among canonical partners ( EML4-ALK, KIF5B-RET, CD74-ROS1 ), the analysis identified a persistent 1-2% rate of non-functional mimics. To resolve this complexity, the system generated definitive outputs for each case, including de novo chimeric DNA and protein sequences, predicted 3D structural models, therapeutic actionability assessments, and an AI-curated summary of comparable literature evidence. Conclusions: Functioning as a comprehensive computational firewall, this framework integrates de novo sequence generation, structural modeling, and therapeutic intelligence to definitively filter out inert genomic noise, ensuring that life-altering TKI prescriptions are grounded in verified functional reality rather than ambiguous terminology.

DEPECA-1 (Defeating Penile Cancer 1): A phase II study to evaluate a first-line systemic therapy with enfortumab vedotin plus avelumab for advanced and metastatic penile carcinoma.

TPS5133 Background: Squamous cell carcinoma of the penis (PeCa) is a rare malignancy with unfavorable outcomes in advanced stage. Since the 1990s, platinum-based chemotherapy has been the standard of care for the first-line treatment of metastatic disease. Unfortunately, it is characterized by a response rate of 30-40%, overall survival (OS) of 17 months and progression free survival (PFS) of 6 months at most. Thus, there is a critical medical need to assess novel systemic strategies for PeCa in the first-line setting, given that current regimens are of a limited clinical benefit while exposing patients to considerable chemotherapy-associated toxicities. In this context, the PD-1 inhibitor pembrolizumab yielded a promising activity in squamous cell carcinomas of various origins. Meanwhile, enfortumab vedotin (EV), a Nectin-4 directed antibody-drug-conjugate, in combination with pembrolizumab has been recently approved for the treatment of metastatic urothelial cancer. The incidence of ≥3 grade adverse events with this regimen was &gt; 55%. Notably, a PD-L1 inhibitor avelumab provided a comparable activity as pembrolizumab but a lower overall rate of endocrine adverse events in patients with urothelial carcinoma. Avelumab monotherapy yielded a response rate of 17% and a median duration of response of 16 months in males with a platinum-refractory PeCa or those unfit for platinum chemotherapy in ALPACA trial. Given that approximately 60% of PeCa tissues express both PD-L1 and Nectin-4, the combination of EV with avelumab represents a potentially synergistic therapeutic strategy, leveraging both direct cytotoxicity and immune-mediated anti-tumor activity through complementary mechanisms of action. Methods: The DEPECA-1 trial is an investigator-initiated, open label, single-arm, multicenter phase II trial, enrolling 25 males with locally advanced or metastatic PeCa at 10 sites in Germany. Patients must be ineligible for curative surgery and not have received any prior systemic palliative therapy. Participants (ECOG ≤ 2) will receive 1200 mg avelumab on day 1 and EV (1,25 mg/kg) on day 1 and day 8 in a 3-week -cycle for a maximum of 24 months and 32 cycles. Primary endpoint is the objective response rate (ORR) assessed per RECIST 1.1. Secondary endpoints include PFS, OS, duration of response (DOR), disease control rate (DCR) and patient-related outcomes. Exploratory analysis comprises tumor biomarker profiling. DEPECA-1 has received regulatory approval on 29 th October 2025. First patient was enrolled on December 16, 2025. Registration IDs: EU CT No. 2025-521644-37-00/NCT07110038. Clinical trial information: NCT07110038 .

Obesity-associated host factors and intracranial disease phenotype in breast cancer brain metastases.

e14030 Background: Obesity is characterized by chronic low-grade systemic inflammation and altered metabolic signaling, which may disrupt periphery-to-brain communication and compromise central nervous system interfaces. These obesity-associated inflammatory perturbations may influence tumor seeding and the phenotypic presentation of brain metastases (BrM). However, the relationship between body mass index (BMI) and intracranial disease phenotype and outcomes in patients with BrM remains poorly defined. Methods: We conducted a retrospective cohort study of patients with breast cancer and BrM with BMI recorded within one year prior to or ≤1 month after BrM diagnosis. The primary endpoint was overall survival (OS) from BrM diagnosis. BMI was analyzed as a continuous variable (per 5 kg/m² increase) and categorically (BMI ≥30 vs &lt;30 kg/m²). Multivariable Cox regression models for OS and intracranial progression-free survival (iPFS) were adjusted for age at BrM diagnosis, receptor-defined subtype (hormone receptor (HR)+/HER2−, HER2+, triple negative breast cancer (TNBC)), Karnofsky performance status, leptomeningeal disease (LMD), intracranial disease burden, local brain-directed therapy, and systemic therapy at BrM diagnosis. Logistic regression evaluated associations between BMI and disease characteristics. In exploratory analyses patients with BMI documented within one year of early breast cancer, time to BrM was evaluated using an accelerated failure time model adjusted for age and subtype. Results: Among 277 patients with BMI at BrM diagnosis, median age was 54.3 years; 109 (41.9%) had HR+/HER2− disease, 84 (32.3%) HER2+, and 67 (25.8%) TNBC. Sixty-four patients (23.3%) were obese (BMI ≥30 kg/m²). Higher BMI was associated with numerically higher intracranial disease burden (adjusted OR per 5 kg/m²: 1.09, 95% CI 0.81–1.78, p=0.677) and lower odds of LMD at presentation (adjusted OR per 5 kg/m²: 0.85, 95% CI 0.56–1.11, p=0.356), although these associations were not statistically significant. BMI was not associated with OS (adjusted HR per 5 kg/m²: 0.99, 95% CI 0.80–1.22, p=0.909) or iPFS (adjusted HR: 0.88, 95% CI 0.70–1.11, p=0.292). When BMI was categorized, moderate and high BMI groups showed numerically shorter OS compared with BMI &lt;25 kg/m², not statistically significant. In exploratory analyses of 142 patients, BMI was not associated with time to BrM among patients with early-stage breast cancer (adjusted time ratio per 5 kg/m²: 1.02, 95% CI 0.91–1.15). Conclusions: Higher BMI at the time of BrM diagnosis was associated with a distinct intracranial disease phenotype, including non-significant differences in disease burden and leptomeningeal involvement, but was not significantly associated with intracranial progression or OS. These findings suggest that host metabolic status may contribute to heterogeneity in BrM and warrants further investigation.

Perioperative toripalimab in non-small cell lung cancer: Clinical outcomes and safety from a large prospective real-world study.

8010 Background: The phase III NEOTORCH trial demonstrated that perioperative toripalimab significantly improved event-free survival in stage IIIA–IIIB driver-gene-negative NSCLC, along with a favorable overall survival trend. However, the generalizability of randomized controlled trial (RCT) results to broader, more heterogeneous, real-world populations is uncertain. This study aimed to evaluate the effectiveness and safety of this regimen in routine clinical practice, providing essential evidence for treatment decisions. Methods: This nationwide, prospective, observational study in China consecutively enrolled patients with stage II-III NSCLC planned for perioperative treatment containing toripalimab. The primary endpoint was real-world event-free survival (rwEFS). Secondary endpoints included pathological complete response (pCR) rate, major pathological response (MPR) rate, objective response rate (ORR), R0 resection rate, real-world disease-free survival (rwDFS), real-world overall survival (rwOS), and safety. Results: Between December 9, 2024, and January 14, 2026, 1727 patients were enrolled from 221 tertiary and secondary hospitals across 28 provinces in China. Baseline characteristics: 82.8% male; 51.4% aged ≥65 years; ECOG PS 0/1/2 in 43.2%/54.4%/2.4%. Histology: 71.4% squamous, 18.3% nonsquamous, 10.3% NSCLC not otherwise specified. Clinical stage distribution: IIA (1.7%), IIB (10.4%), IIIA (52.8%), IIIB (33.0%), IIIC (2.1%). Of all enrolled patients, 542 (31.4%) were pending preoperative assessment. Surgery was not performed in 634 patients (36.7%); reasons included: other causes (64.9%), patient refusal (15.7%), surgical ineligibility (9.3%), disease progression (8.2%), and adverse events (1.9%). Ultimately, 551 patients (31.9%) underwent resection, with 503 having postoperative pathological assessment. The MPR rate was 67.0% (337/503) and the pCR rate was 39.6% (199/503). As of the data-cutoff, investigators reported 20 (1.2%) grade ≥3 or clinically significant treatment-emergent adverse events, with hematologic toxicities being most common. Immune-related adverse events were infrequent (four events: three grade 1–2 pneumonitis, one grade 3 rash). Conclusions: This is the largest prospective real-world study of perioperative immunotherapy in NSCLC to date. Its extensive geographic coverage and diverse patient population-including higher proportions of elderly patients, those with ECOG PS 1, and varied histologies-validate the effectiveness and manageable safety of perioperative toripalimab in routine practice beyond RCT settings. The significant pathological response rates observed support the clinical benefit of this regimen for a broad spectrum of stage II-III NSCLC patients, providing crucial evidence for real-world decision-making. Clinical trial information: ChiCTR2400091457.