Browse Articles

An optimized machine learning model for overall survival prediction in brain metastasis patients using genomic mutation and copy number features.

e14002 Background: Cancer progression and patient survival are influenced by both tumor-intrinsic and microenvironmental factors, including the ability of tumor cells to disseminate and colonize distant organs. Organ-specific metastases, such as brain metastases, exhibit distinct tumor–microenvironment interactions, therapeutic responses, and clinical outcomes. Incorporating metastatic genomic patterns into survival modeling is critical for improving prognostic accuracy. Here, we present an optimized machine learning framework using genomic mutations and copy number variations to predict overall survival (OS) in brain metastasis (BM) patients. Methods: We employed a rigorous, machine learning methodology to build a survival prediction model. Feature selection, model training, and hyperparameter optimization were conducted exclusively within the training dataset, while the test dataset was held out for final evaluation. The cohort was randomly split into training (70%) and test (30%) datasets. Prognostic features were first identified in the training cohort using univariable Cox regression (p < 0.05) and further refined using machine learning–based feature selection, retaining features selected by at least 12 models. Hyperparameter tuning was performed using 3-fold cross-validation. The Model was assessed using the concordance index (C-index) and area under the curve (AUC), and survival analysis was conducted using the Kaplan–Meier method. Results: The cohort included 381 patients with brain metastases primarily from lung cancer (51.1%), followed by melanoma (15.2%) and breast cancer (7.9%). Among the selected prognostic features, many were single-nucleotide variants, with alterations in PTPRT, ARID1A, PREX2, and FAT1 frequently represented across metastatic malignancies. Ridge regression emerged as the top-performing model, achieving a C-index of 0.70 in the test cohort, demonstrating robust performance. As summarized in Table 1, time-dependent and survival analyses indicate stable model performance, with consistent discrimination over time and clear separation of predicted risk groups in the internal test cohort. Conclusions: We developed an optimized machine learning framework that integrates genomic mutation profiles and copy number alterations to predict overall survival in patients with brain metastases. The model demonstrated robust and stable performance across time-dependent and survival analyses, achieving clear risk stratification in an independent test cohort, thereby supporting its potential clinical utility for prognostic risk assessment in the clinical settings. Summarized model performance. Metric Time/Comparison Value Time-dependent AUC year 1 0.642 Time-dependent AUC year 2 0.711 Time-dependent AUC year 3 0.729 Kaplan–Meier HR High vs Low risk 3.45 p-value High vs Low risk < 0.001

Trastuzumab deruxtecan (T-DXd) for pretreated patients in China with HER2 IHC 3+ solid tumors: DESTINY-PanTumor03 Part 1 primary analysis.

3026 Background: Treatment options in China are limited for patients with late-line, HER2-expressing advanced solid tumors. In Part 1 of DESTINY-PanTumor02, T-DXd showed clinically meaningful antitumor activity in HER2-expressing advanced solid tumors, with the greatest benefit observed in HER2 IHC 3+ tumors. Based in part on these findings, T-DXd has been approved in multiple countries worldwide, including the US, as treatment for patients with unresectable or metastatic HER2-positive (IHC 3+) solid tumors who have received prior treatment and/or have no satisfactory alternative therapies. Following the results from DESTINY-PanTumor02, DESTINY-PanTumor03 is evaluating T-DXd in patients in China with HER2-expressing advanced solid tumors. The primary analysis of Part 1 of DESTINY-PanTumor03 is presented here. Methods: DESTINY-PanTumor03 is an open-label, Phase 2 study (NCT06271837). Part 1 is evaluating T-DXd (5.4 mg/kg IV Q3W) in patients in China with HER2 IHC 3+ (by central testing), locally advanced, unresectable, or metastatic solid tumors (excluding breast and gastric cancers) after ≥1 prior systemic treatment for advanced disease or without treatment options. The primary endpoint is confirmed objective response rate (ORR) by independent central review (ICR) per RECIST 1.1. Secondary endpoints include ORR by investigator assessment (INV) per RECIST 1.1; duration of response (DOR), disease control rate (DCR), and progression-free survival (PFS) by INV and ICR per RECIST 1.1; overall survival (OS); and safety. Results: At primary analysis data cutoff (November 28, 2025), 50 patients with biliary tract (n = 11), colorectal (n = 6), cervical (n = 10), endometrial (n = 7), ovarian (n = 5), non-small cell lung (n = 7), or other cancers (n = 4) had received T-DXd. Median follow-up duration was 9.9 (range 1.1–18.3) months. Median number of prior treatment regimens was 2 (range 1–10). By ICR, ORR (95% CI) was 58.0% (43.2, 71.8), median DOR (95% CI) was 15.4 (12.5, not evaluable [NE]) months, DCR (95% CI) at Week 6 was 88.0% (75.7, 95.5), and median PFS (95% CI) was 15.7 (7.2, NE) months. By INV, ORR (95% CI) was 56.0% (41.3, 70.0). Median OS was not reached. Grade ≥3 drug-related adverse events occurred in 31 (62.0%) patients, and adjudicated drug-related interstitial lung disease / pneumonitis occurred in 4 (8.0%) patients (Grade 2 n = 3 [6.0%], Grade 3 n = 1 [2.0%]). Conclusions: T-DXd demonstrated durable and clinically meaningful antitumor activity in pretreated patients in China with HER2 IHC 3+ advanced solid tumors. Safety was generally consistent with the established T-DXd profile. Results from DESTINY-PanTumor03 Part 1 support T-DXd as a tumor-agnostic treatment for patients in China with HER2 IHC 3+ solid tumors. Clinical trial information: NCT06271837 .

Economic evaluation of sterilization reversal in infertility treatment: A systematic review

Objectives Although sterilization is intended to be permanent, some individuals later seek fertility. In such cases, options can be limited and financially burdensome. This review evaluated the cost-effectiveness of sterilization reversal surgery in previously sterilized individuals. Methods We searched MEDLINE, Embase, and Scopus from inception in 1946 through December 2025, following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidance. We included studies that analyzed cost-effectiveness or reported the costs of sterilization reversal in males (vasectomy reversal) or females (tubal anastomosis), with assisted reproductive technologies as comparators. Study quality was assessed using the Consolidated Health Economic Evaluation Reporting Standards 2022 checklist. Two authors independently screened each study to reduce bias. All costs per outcome were converted to 2024 United States dollars for analysis and comparisons. Results Of 1628 identified articles, 24 studies met the eligibility criteria. Almost all examined populations in high-income countries, such as the United States, the Netherlands, and Singapore. Thirteen studies evaluated tubal anastomosis, and eleven evaluated vasectomy reversal. Most studies reported lower total costs for sterilization reversal than for assisted reproductive technologies, with comparable outcomes. Vasectomy reversal was preferred for male patients irrespective of the female partner’s age, whereas tubal anastomosis was preferred for female patients aged 40 years or younger. For older patients, assisted reproductive technologies were more cost-effective. Conclusions Tubal anastomosis and vasectomy reversal may be economically advantageous compared with assisted reproductive technologies for infertility due to prior sterilization. However, societal factors, including a country’s socioeconomic context and policy feasibility, should be considered.

Recent advances in high surface area porous materials for electrochemical energy storage devices

Multi‐Field Synergy for Orchestrating Filler Angles in Polyimide Aerogels with Switchable Electromagnetic Interference Shielding

ABSTRACT The fast‐evolving IT sector necessitates intelligent electromagnetic interference (EMI) shielding materials capable of real‐time, environment‐responsive. While current approaches based on reconstructing conductive networks through mechanical strain enable dynamically responsive shielding, but face a narrow tuning range, inadequate stability, and practical limitations. To address this, we propose an electric/magnetic field synergistic regulation strategy. This approach enables precise control over the alignment angle between reduced graphene oxide (rGO) and nickel nanowires (NiNWs) by manipulating the external field direction, producing rGO@NiNWs/polyimide aerogels with 3D ordered networks. Leveraging this design, the aerogels achieve reversible, wide‐range tuning of EMI shielding performance through simple physical rotation, enabling reliable “on/off” switching capability. The oriented structure also optimizes both filler interconnection efficiency and interfacial polarization. With an rGO@NiNWs content of 80 wt.% and an inter‐phase angle of 90°, the aerogels demonstrate excellent ultra‐wideband EMI shielding performance across gigahertz and terahertz bands, with an average shielding effectiveness of 85 dB in the terahertz band, alongside good stability in extreme environments. Finite element simulations further reveal how the spatial configuration of rGO@NiNWs governs the shielding behavior and intelligent response mechanism. This study paves the way for next‐generation intelligent electromagnetic protection materials, with promising potential for aerospace and wearable applications.

Knowledge graph-enhanced heterogeneous graph neural network for scientific talent innovation potential identification

Abstract Traditional talent evaluation methods predominantly rely on static bibliometric indicators that fail to capture the dynamic evolution patterns and potential innovative capabilities of researchers. This study proposes a novel knowledge graph-enhanced heterogeneous graph neural network framework for identifying innovation potential in scientific talents. The framework integrates multi-source heterogeneous academic data to construct a comprehensive knowledge graph encompassing researchers, publications, institutions, and research topics, while employing meta-path-based attention mechanisms to selectively aggregate information from diverse entity types and relationships. A gated fusion strategy adaptively combines semantic embeddings from knowledge graphs with structural features from academic networks, enabling comprehensive talent representation learning. Experimental validation on a dataset containing 128,456 researchers across multiple disciplines demonstrates superior performance, achieving 85.21% accuracy and 0.9014 AUC-ROC score, representing significant improvements of 6.3% over state-of-the-art baseline models. The proposed approach exhibits particular effectiveness in identifying early-career researchers with high innovation potential, addressing cold-start problems inherent in conventional evaluation systems. This research provides a generalizable methodology for knowledge-augmented graph representation learning and offers practical solutions for intelligent talent management in research institutions.

Mapping the multi-domain allosteric network for CKAMP44 modulation of AMPA receptors

Association of GFPT2 expression with clinical outcomes in clear cell renal cell carcinoma.

e16543 Background: Progression of clear cell renal cell carcinoma (ccRCC) is driven in part by tumor microenvironment remodeling, including epithelial mesenchymal transition (EMT), stromal activation, and metabolic reprogramming. GFPT2, the rate-limiting enzyme of the hexosamine biosynthesis pathway, has been linked to extracellular matrix remodeling and EMT across cancers and is enriched in high-grade, advanced ccRCC. We hypothesized that GFPT2 expression identifies a microenvironment-associated program linked to adverse clinical outcomes in ccRCC. Methods: We analyzed mRNA expression and clinical outcomes in TCGA-KIRC (n = 498). Survival analyses employed Kaplan-Meier methods and multivariable Cox proportional hazards regression. Pearson correlations quantified GFPT2-microenvironment associations. Outcomes included overall survival (OS) and progression-free survival (PFS). Results: High GFPT2 expression (median cutoff) was associated with significantly worse OS and PFS (log-rank p < 0.001 for both). As a continuous variable, GFPT2 expression conferred markedly inferior OS (HR 1.74, 95% CI 1.49-2.04, p < 0.001) and PFS (HR 1.96, 95% CI 1.66-2.31, p < 0.001) in univariate analysis, with sustained independent prognostic significance after clinical adjustment (OS: HR 1.42, p < 0.001; PFS: HR 1.42, p < 0.001). GFPT2 expression demonstrated robust correlations with EMT effectors (VIM r = 0.29, FN1 r = 0.41, TWIST1 r = 0.29, SNAI1 r = 0.23), stromal collagens (COL1A1 r = 0.51, COL3A1 r = 0.40), and hypoxia signaling (Buffa score r = 0.45). In multivariable models incorporating EMT and stromal markers alongside clinical covariates, GFPT2 retained independent prognostic significance for OS (HR 1.26, 95% CI 1.02-1.54, p = 0.028) with borderline significance for PFS (HR 1.22, 95% CI 0.99-1.49, p = 0.057), indicating partial but incomplete mediation of GFPT2's adverse prognostic impact through microenvironment-associated transcriptional programs. Conclusions: GFPT2 expression independently predicts adverse survival outcomes in ccRCC through mechanisms partially, but not entirely, mediated by EMT- and stroma-related transcriptional reprogramming. These findings position GFPT2 as a candidate biomarker integrating metabolic dysregulation with microenvironment remodeling and warrant investigation as a therapeutic target in high-risk ccRCC. Multivariable Cox Model for Overall Survival (OS) and Progression-Free Survival (PFS). Variable OS: HR (95% CI) p-value PFS: HR (95% CI) p-value GFPT2 1.74 (1.49-2.04) <0.001 1.96 (1.66-2.31) <0.001 GFPT2* 1.42 (1.21-1.65) <0.001 1.42 (1.21-1.66) <0.001 GFPT2§ 1.26 (1.02-1.54) 0.028 1.22 (0.99-1.49) 0.057 GFPT2† 1.37 (1.15-1.64) <0.001 1.37 (1.15-1.64) <0.001 *Adjusted for Age, Sex, and Stage. §Adjusted for Age, Sex, and Stage, and EMT and stromal markers. †Adjusted for Age, Sex, and Stage, and EMT and BUFFA Hypoxia score.

Optimal adjuvant therapy for early-stage cervical cancer (ESCC) with intermediate-risk factors: A target trial emulation of GOG-0263 based on the NCDB.

5539 Background: Optimal adjuvant therapy for early-stage cervical cancer (ESCC) patients with intermediate-risk factors remains uncertain, specifically whether the addition of concurrent chemotherapy (CRT) to radiation therapy (RT) confers a survival benefit. The GOG-0263 randomized phase III trial compared adjuvant RT vs CRT in patients with FIGO Stage I-IIA cervical cancer who had undergone radical hysterectomy. We sought to emulate this trial using real-world data from the National Cancer Database (NCDB) to evaluate the benefit of CRT versus RT on overall survival (OS) in intermediate risk ESCC. Methods: Using the NCDB, we conducted a target trial emulation of GOG-0263 for patients diagnosed from 2004-2020 with pathologic T1-T2a2 N0 M0 cervical cancer who underwent radical hysterectomy. Eligible patients had intermediate risk factors, defined as tumor size ≥ 2 cm with lymph-vascular invasion (LVI) or tumor size ≥ 4 cm. OS, defined from the date of definitive surgery, was compared between patients who received adjuvant RT alone versus adjuvant CRT. Kaplan-Meier methods and Cox proportional hazards models, along with an inverse probability treatment weighting (IPTW) schema, were implemented to assess the treatment effect. Subgroup and sensitivity analyses were conducted to identify specific disease characteristics that maximize the benefit of CRT and to evaluate the impact of potential immortal time bias. Results: Among 1,171 eligible patients, 537 (45.9%) received CRT and 634 (54.1%) received RT alone, with median ages of 44 years (IQR: 36-55) and 46 years (IQR: 39-57) and median follow-up times of 7.5 years (IQR: 4.8-10.5) and 6.6 years (IQR: 4.2-9.9), respectively. There was no significant difference in OS between CRT and RT in unadjusted (HR 0.94; 95% CI, 0.71–1.25), adjusted (HR 0.89; 95% CI, 0.67–1.18), or weighted analyses (HR 0.88; 95% CI, 0.66–1.17). Five-year OS was similar between groups (86.4% vs. 86.6%). No subgroup demonstrated a significant survival benefit from CRT. Conclusions: In this real-world emulation of GOG-0263, addition of concurrent chemotherapy to adjuvant RT was not associated with improved OS for intermediate-risk ESCC. These findings support emerging trial evidence and supplement the conclusion of the GOG-0263 trial. The indication of this study suggests that chemotherapy may be safely omitted in this population under a doctor’s careful discretion, thereby potentially reducing treatment-related toxicity and preserving patients’ quality of life without compromising survival. These findings highlight the need for improved risk stratification and patient selection to identify those most likely to benefit from treatment intensification with concurrent chemotherapy in addition to adjuvant RT.

Integrating HER2-targeted antibody-drug conjugates into mucinous ovarian cancer treatment: A preclinical assesstment.

e17545 Background: Mucinous ovarian cancer (MOC) is a rare epithelial ovarian cancer. There are currently no systematic therapies in clinical use for MOC that reliably improve patient outcome, therefore new approaches are needed. MOC genetic alterations and the resulting proteins are candidates for targeted molecular therapies. About a quarter of MOC carry high-level ERBB2 (HER2) gene amplification, and early case reports were promising for anti-HER2 targeted agents such as trastuzumab. We hypothesized that the ADC trastuzumab deruxtecan (T-DXd) will show efficacy in models of MOC expressing HER2. Methods: HER2 expression was assessed by immunohistochemistry in MOC tumours, organoids, and xenografts. ERBB2 amplification and model fidelity were confirmed using whole genome sequencing and STR profiling. The efficacy of T-DM1 and T-DXd was tested in vitro using organoid models with varied HER2 expression (3+, 2+/1+, 0), cultured in reduced Matrigel to enhance drug penetration. In vivo efficacy of T-DXd is being evaluated in xenograft models (Org49: 3+, Org60: 1+/2+). Results: HER2 expression was heterogeneous across MOC models, with 2 of 12 organoid lines showing strong (3+) IHC staining and 2 displaying moderate expression. Whole genome sequencing confirmed ERBB2 amplification in two organoids line, correlating with high HER2 protein levels. In vitro testing demonstrated greater efficacy of T-DXd compared to T-DM1 in HER2-positive models, with little to no response in HER2-negative controls. Optimizing 3D culture conditions by reducing Matrigel concentration improved ADC penetration and response. In vivo validation using PDX models (Org49, Org60) is currently underway to assess therapeutic potential. Conclusions: HER2 is a promising therapeutic target in a molecularly defined subset of MOC. T-DXd demonstrated strong preclinical activity in HER2-positive MOC models and warrants further investigation in vivo to support future clinical translation.

Clinical trial participation among adolescents and young adults diagnosed with cancer in the United States (2004-2021).

1560 Background: Adolescents and young adults (AYAs, 15-39 years) have low historical participation in cancer clinical trials (CTs) than pediatric patients, which limits access to novel therapies and slows progress in outcomes. Although cooperative-group networks and AYA-focused initiatives have improved accrual, enrollment remains uneven across care settings and patient subgroups, highlighting the need for contemporary population-level estimates of participation and its predictors. Methods: The National Cancer Database was utilized to identify AYAs diagnosed between 2004 and 2021 with one of the 20 cancer types associated with the highest mortality. Patients with confirmed malignant neoplasms, known vital status, and ≥6 months of follow-up were included. The final analytic cohort comprised 865,923 patients. The primary endpoint was documented CT participation (yes/no). Multivariable logistic regression was used to estimate adjusted odds ratio (aOR) of enrollment by diagnosis epoch, age group, sex, race/ethnicity, insurance status, area-level education and median income, residence, Charlson-Deyo comorbidity score, and receipt of surgery, radiation, and systemic therapy. Results: Overall, 2,621 patients (0.3%) were enrolled in a CT. Enrollment increased from 0.1% (2004–2009) to 0.2% (2010–2013), 0.3% (2014–2017), and 0.5% (2018–2021); compared with 2004–2009, enrollment was higher in 2018–2021 (aOR: 4.92, 95% CI: 4.34–5.60). Compared to those aged <24 years, the odds of enrollment were lower for ages 25–29 (aOR: 0.52, 95% CI 0.46–0.58), 30–34 (aOR: 0.45, 95% CI 0.41–0.51), and 35–39 (aOR: 0.46, 95% CI 0.41–0.50). Enrollment was lower among non-Hispanic Black (aOR: 0.84, 95% CI 0.74–0.96) and Hispanic patients (aOR: 0.79, 95% CI: 0.69–0.90) compared to non-Hispanic White patients. Compared with uninsured patients, enrollment was higher among those with private (aOR: 1.74, 95% CI 1.38–2.21) or government insurance (aOR: 1.60, 95% CI: 1.26–2.05), and was also higher in areas with greater education attainment (aOR: 1.14, 95% CI: 1.03–1.25) and individuals in the high median income group (aOR: 1.13, 95% CI: 1.02–1.25). Patients who had received radiation (aOR 1.56, 95% CI 1.44–1.70) or systemic therapy (aOR 3.59, 95% CI 3.23–4.01) were more likely to enroll. Conclusions: CT participation among AYAs was exceedingly low but has increased over time. Differences in enrollment by age, race/ethnicity, insurance, and area-level socioeconomic measures suggest persistent barriers to CT access and participation. These findings support expanding trial availability beyond high-resource settings, strengthening AYA-focused referral pathways and infrastructure, and implementing equity-centered strategies such as navigation, reduced trial burden, and financial/insurance support to improve representation and access to investigational therapies.

Safety, pharmacokinetic, and preliminary efficacy of SNC115 in patients with relapsed/refractory small cell lung cancer (SCLC) and large cell neuroendocrine carcinoma (LCNEC): A phase 1 study.

8106 Background: SCLC and LCNEC are characterized by aggressive clinical courses and limited therapeutic options following progression on platinum-based chemotherapy. DLL3(Delta-like ligand 3) is highly expressed and selectively on the surface of these high-grade neuroendocrine tumors, making it an ideal therapeutic target. SNC115 is a novel CAR-T cell armored with CD70 to target DLL3-expressing cells. Methods: This phase 1, open-label study utilized an accelerated titration design followed by a standard 3+3 escalation to evaluate the safety, tolerability, and pharmacokinetics of SNC115 in patients (pts) with R/R SCLC or LCNEC who progressed after ≥1 line of standard therapy. Following leukapheresis and a 3-day lymphodepletion regimen (fludarabine/cyclophosphamide), pts received a single infusion of SNC115 at one of five planned dose levels (DL): 1.0×10^5 (DL1), 3.0×10^5 (DL2), 1.0×10^6 (DL3), 3.0×10^6 (DL4), and 6.0×10^6 (DL5) CAR + T cells/kg. The primary endpoints were the determination of the maximum tolerated dose (MTD) and the recommended dose (RD). Results: As of 23 January 2026, 8 pts with R/R SCLC were treated across four DLs: DL1 (n = 1), DL2 (n = 3), DL3 (n = 2) and DL4 (n = 2). Median age was 53.5 (range 40-69) years, with a median of three prior therapy lines (range 2-6). 6 pts received bridging therapy. SNC115 demonstrated a manageable safety profile. CRS occurred in 3 pts (DL1: 1pt, DL4: 2pts), all Grade 1 or 2 and resolved with tocilizumab/corticosteroids. Grade ≥3 hematologic TEAEs included lymphocyte count decreased (8/8), white blood cell count decreased (2/8) and anaemia (1/8), all attributed to LD. Only one patient (DL4) experienced Grade ≥3 non-hematologic TEAEs related to SNC115, including alanine aminotransferase increased (G3), Gamma-glutamyltransferase increased (G3) and diarrhea (G3), which resolved with symptomatic care. No DLTs, SAEs or ICANS were reported. Among evaluable pts (n = 8), the ORR and DCR were 37.5% and 75%, respectively. All patients (n = 2) in the higher-dose level (DL4) achieved PR. CAR-T cells expansion peaked at a median of 7 days post-infusion, with Cmax ranging from 68.77 to 4359.89 copies/µg DNA. Conclusions: SNC115 demonstrated a manageable safety profile and encouraging preliminary antitumor activity in heavily pretreated pts with SCLC. The absence of DLTs and ICANS, combined with low-grade transient CRS, supports continued dose escalation. These early efficacy signals suggest that SNC115 may provide a novel therapeutic avenue for R/R SCLC. Clinical trial information: NCT06384482 .

A phase 1b study of tegavivint, a TBL1 inhibitor, with gemcitabine in patients with relapsed or refractory osteosarcoma (TIGER).

TPS11586 Background: The Wnt-signaling pathway and its downstream transcriptional effector, β-catenin, are dysregulated in osteosarcoma (OS), promoting tumor growth and metastatic spread. When the Wnt pathway is activated, transduction beta-like protein 1 (TBL1) binds nuclear β-catenin, forming a complex that is necessary to dock on the promoters of Wnt-driven genes. Tegavivint is a first-in-class small molecule inhibitor that selectively binds TBL1, inhibiting TBL1-β-catenin complex formation, and resulting in the nuclear degradation of β-catenin and inhibition of oncogenic transcriptional activation. Preclinical testing of in vivo OS models demonstrates that tegavivint reduces tumor volume and suppresses lung metastases, and has additive effects when combined with gemcitabine. In a phase 1 trial conducted by the Children’s Oncology Group, the recommended phase 2 dose (RP2D) of tegavivint was determined to be 6.5 mg/kg administered intravenously (IV) on a 3 weeks on/1 week off schedule, with no maximum tolerated dose (MTD) determined. The primary objective of this phase 1b study is to define the MTD and/or RP2D of tegavivint combined with gemcitabine, including assessment of toxicities and preliminary efficacy, in patients with relapsed or refractory (r/r) OS. Methods: TIGER is a phase 1b, multi-center, dose escalation trial assessing tegavivint combined with gemcitabine in patients with r/r OS. Eligible patients will be 1-30 years of age with either measurable or evaluable disease per RECIST v1.1 who have fully recovered from the clinically significant acute effects of prior therapy and have adequate organ function. Prior treatment with gemcitabine is allowed. Patients with active CNS disease, recent bisphosphonate treatment, metabolic bone disease or disorder, uncorrected hypocalcemia or low vitamin D, and prior receipt of tegavivint are excluded. Patients will receive tegavivint IV at the dose level assigned at study entry on days 1, 8, and 15 and gemcitabine at a fixed dose of 1000 mg/m 2 IV on days 1 and 8 of a 21-day cycle. Patients may receive up to 17 cycles provided they do not experience disease progression or unacceptable toxicity. Tegavivint will be dose escalated using a rolling six design to test two planned dose levels, 5 mg/kg [dose level (DL) 1] and 6.5 mg/kg (DL 2), with dose de-escalation to 3 mg/kg (DL 0) if needed. An additional 6 patients will be enrolled to a dose expansion cohort at the MTD/RP2D to obtain additional pharmacokinetic (PK), safety, and preliminary efficacy data. Peripheral blood for correlative studies will be obtained at serial timepoints to conduct PK studies, pharmacodynamic testing of serum protein biomarkers associated with β-catenin inhibition, and analysis of ctDNA and circulating tumor cells. Tumor tissue will be analyzed for TBL1 and Wnt/β-catenin signaling (sequencing and protein expression). Clinical trial information: NCT07144254 .

A phase II, single-arm trial of adebrelimab plus nab-paclitaxel and carboplatin as first-line therapy for advanced thymic carcinoma: Interim efficacy and safety results.

e20171 Background: Thymic carcinoma (TC) is a rare and aggressive malignancy with poor prognosis in advanced stages. Given the limited efficacy of standard first-line regimen (carboplatin/paclitaxel), there remains an urgent unmet need for novel treatment approaches. This study evaluated the efficacy and safety of the PD-L1 inhibitor adebrelimab in combination with nanoparticle albumin-bound (nab)-paclitaxel and carboplatin as first-line therapy for advanced or recurrent TC. Methods: Patients (pts) with unresectable UICC stage III or IV, recurrent, or metastatic TCs without any previous anti-tumor therapy were enrolled.During the induction phase, pts received adebrelimab (20 mg/kg) plus nab-paclitaxel (260 mg/m 2 ) and carboplatin (AUC 5) every 3 weeks for up to 4-6 cycles. This was followed by the maintenance phase, adebrelimab (20 mg/kg) every 3 weeks for up to 2 years (including the induction phase), until progression or unacceptable toxicity. The primary endpoint was objective response rate (ORR), and secondary endpoints included progression free survival (PFS), disease control rate (DCR), overall survival (OS), and safety. Results: Between August 15, 2023, and June 3, 2025, this study was conducted at two centers and enrolled a total of 37 pts. All pts were included in safety and efficacy analysis. The median age was 60.0 years old (range 29-71). A total of 26 pts (70.3%) pts completed the induction therapy and were transitioned to maintenance therapy while 11 pts (29.7%) remained on treatment at data cutoff (December 1, 2025). Four (10.8%) of 37 pts had complete response, 16 (43.2%) had partial response, and 17 (45.9%) had stable disease. The ORR was 54.1% (20/37) and DCR was 100% (37/37). The median PFS was 10.9 months (95% CI, 9.1-NA). Treatment-related adverse events (TRAEs) of any grade and of grade ≥3 severity occurred in 100% and 54.1% (20/37) of pts, respectively. The most common TRAEs were anemia, lymphocyte count decreased, and white blood cell count decreased. The immune-related AEs of grade ≥3 severity occurred in 16.2% (6/37) of pts, including immune-mediated rash, myositis and amylase increased. No grade 5 TRAEs were reported. Conclusions: Adebrelimab combined with nab-paclitaxel and carboplatin demonstrates promising efficacy and a manageable safety profile as first-line therapy for advanced or recurrent thymic carcinoma, offering a new potential treatment option for this patient population. Trial Registration: ChiCTR2300072705.

Real-world characterization of SEZ6, a transmembrane protein expressed in various solid tumors.

3082 Background: Seizure-related homolog 6 (SEZ6) is a transmembrane protein involved in neuronal development that is downregulated in normal adult tissues outside the central nervous system (CNS). It is overexpressed in various solid tumors such as small cell lung cancer (SCLC), neuroendocrine neoplasms (NENs), and CNS tumors, with strong correlations between mRNA and protein expression. In SCLC, limited data suggest that SEZ6 expression is greater in molecular subtypes with neuroendocrine (NE)-high gene signatures (i.e., SCLC-A and SCLC-N) than in non-NE subtypes (i.e., SCLC-P). However, SEZ6 expression in tumor tissues remains poorly understood and existing data are limited by small sample sizes. This study aimed to characterize SEZ6 expression in a large patient population, across multiple tumor types. Methods: Pan-tumor data from a real-world database (Tempus AI, Inc.) were analyzed for SEZ6 mRNA expression in 37,645 samples, with a focus on lung, brain, and NE tumor types. In SCLC, SEZ6 expression was compared between SCLC-A, -N, -P, and -I samples via hierarchical clustering. Gene expression was reported as log 2 (transcripts per million [TPM]+1). Results: Across the 37,645 samples analyzed, tumor types with the highest median (log 2 [ SEZ6 TPM+1]) expression were astrocytoma (5.8; n=354), glioma (5.8; n=502), NENs (5.7; n=2794), and glioblastoma (5.6, n=2833). Among NENs, the highest SEZ6 expression was seen in prostate neuroendocrine carcinoma (NEC) (6.6; n=176), SCLC (6.5; n=2018), and bladder small cell NEC (6.5; n=139). Median SEZ6 expression was 3–45-fold higher than other biomarkers of interest in SCLC ( B7-H3 , 3-fold; DLL3 , 3-fold; CD274 [PD-L1], 28-fold; PDCD1 [PD-1], 45-fold). In SCLC samples, median SEZ6 expression was stable across primary (6.4) and metastatic (6.5) tumors and stage II to IV disease (6.6–6.5). Among SCLC molecular subtypes, median SEZ6 expression was lower in non-NE SCLC-P tumors (3.8; n=85) than in SCLC-I (6.0; n=175), or in NE-high SCLC-N (6.7; n=339) and SCLC-A (7.3; n=224) tumors. SEZ6 expression was 6–7-fold higher in prostate NEC than in non-NE metastatic prostate adenocarcinoma (castration resistant, 6-fold; castration sensitive, 7-fold). Conclusions: SEZ6 is highly expressed in SCLC and tumors of NE origin, particularly prostate and bladder NEC. In SCLC, SEZ6 expression is higher than other targets of antibody–drug conjugates (ADCs) approved or in development for SCLC, supporting SEZ6 as a promising therapeutic target. SEZ6 expression is highest in SCLC-A and SCLC-N molecular subtypes. High SEZ6 expression is also seen in SCLC-I, a subtype associated with platinum resistance, suggesting SEZ6 may be a potential therapeutic target in certain patients with platinum-refractory SCLC. Taken together, these results suggest cross-indication applicability of targeting SEZ6, highlighting its potential as a multi-tumor ADC target warranting further investigation.

Travel time to cancer care as it relates to care access, transportation distress, and financial toxicity: Findings from the Cancer Experience Registry.

1537 Background: Longer travel time to cancer care has been linked to worse prognosis and quality of life for patients. This study aims to examine how travel time to cancer care is associated with access to care, transportation-related distress, and financial toxicity. Methods: Participants in Cancer Support Community’s Cancer Experience Registry (CER) from Oct 2021-Dec 2025 reported sociodemographic and clinical information, financial toxicity (11-item FACIT-COST), travel time to cancer care (min/hr), transportation-related distress (1= not at all to 5= very seriously concerned ), and delays/barriers to accessing cancer care (yes/no). Descriptive findings and between-groups analyses (Chi-square) are presented. Results: 3,590 U.S. adults (75% women; 82% non-Hispanic (NH) White, 7% NH Black, 5% Hispanic/Latino, 5% multiple or other races; 18-95 years old [Mean=62.2, SD=13.1]) reported various cancer diagnoses (34% breast, 25% blood, 10% colorectal, 7% gynecologic, 25% other), with 54% in remission/NED, 18% localized, and 15% metastatic (time since diagnosis Median=3yrs). FACIT-COST indicates 52% with no financial toxicity, 26% mild, 20% moderate, and 2% severe financial toxicity (M=24.9, SD=12.0). For travel time, 67% of participants reported < 1hr, 23% 1–2hrs, 7% 3–4hrs, and 4% 5+hrs. 79% reported no transportation distress, while 10% were slightly concerned and 12% moderately to very seriously concerned. Over one-quarter (29%) reported ever experiencing a delay in cancer care. Among patients traveling <1hr to their care center, 24% reported experiencing delays/barriers, compared with 46% of those traveling 5+hrs (χ²=84.3, p<.001). Similarly, transportation-related distress increased with travel time: 84% of patients traveling <1hr reported no concern, whereas only 60% of those traveling 5+hrs reported no concern, with 29% reporting moderate to very serious concern (χ²=167.9, p<.001). Lastly, financial toxicity increased with travel burden: 43% of patients traveling <1hr reported mild-to-severe financial toxicity versus 64% of those traveling 5+hrs (χ²=63.7, p<0.001), and 85% of those reporting moderate to severe transportation distress also reported at least mild financial toxicity (χ²=388.4, p<0.001). Conclusions: Most patients surveyed reported living within 1hr of their care site and experienced little transportation distress; however, those who travel longer distances for care report higher rates of delays/barriers, greater transportation-related distress, and more financial toxicity. Collectively, these findings underscore that longer travel times are linked to both objective barriers and subjective distress, highlighting that travel burden may limit timely cancer care and worsen patient outcomes. Policy solutions that bring quality care closer to patients are essential to reduce travel burdens and distress.

Post–PD-1 failure in advanced melanoma: Real-world outcomes with lenvatinib-based versus non-lenvatinib-based therapy.

e21520 Background: Post–anti–PD-1 advanced melanoma has few effective salvage options and CHEMO outcomes are modest. We evaluated real-world comparative effectiveness (incl. OS) of LENVA+Pembro versus non-Lenva-based therapy after anti–PD-1 progression, given encouraging activity in LEAP-004 despite negative 1L data in LEAP-003. Methods: Single-center retrospective RWD study at N.N. Blokhin NMRCO using the full EMR (01/2022–09/2025; 37,942 records; 7,061 pts). Advanced cutaneous melanoma pts progressing after anti–PD-1 were identified (ICD-10 C43 + free-text; non-cutaneous primaries manually excluded). Results: Treatment assignment was available for 235 patients (chemo without lenva, n = 117; lenva-based regimen, n = 118). Baseline characteristics were broadly similar: median age 59 (32–95) vs 61 (25–86) years, men 46.2% vs 37.3%, BRAF-mutant 30.8% vs 28.0%, and metastatic stage at treatment start (M1d 26.5% vs 22.9%; M1c 32.5% vs 32.2%). Prior ipilimumab was more frequent in the lenva group (66.1% vs 56.4%). The main imbalance was the therapy line: among patients with documented lines (70/117 and 80/118), the median line was 2 (1–6) vs 2.5 (1–6) and ≥3rd line used 42.9% vs 50.0%. Median FU was 16.2 mo (95% CI 11.2–22.2) with chemo and 14.4 mo (95% CI 10.8–22.6) with lenva. In OS Kaplan–Meier analysis (N = 235; 82 deaths), median OS was 77.2 mo (95% CI 24.5–NR) with chemo (37 deaths) and 24.6 mo (95% CI 13.4–NR) with lenva-based therapy (45 deaths) (log-rank p = 0.183). A multivariable Cox model with time-varying lenva effect was fitted in pts with known BRAF (WT/mutant) and complete covariates (n = 129; 46 deaths), adjusting for age, BRAF, metastatic stage at treatment start, prior ipilimumab, and line of therapy. Line of therapy was independently associated with OS (HR 1.36 per one-line increase; 95% CI 1.03–1.79). To account for non-proportional effects, lenva exposure was modeled piecewise (0–6, 6–12, ≥12 months), yielding interval-specific hazard ratios vs chemotherapy-only: 0–6 mo HR 0.82 (95% CI 0.35–1.92), 6–12 mo HR 3.13 (0.65–14.99), and ≥12 mo HR 3.11 (0.80–12.09). The proportional hazards test for the interval-specific lenva effect was borderline (p = 0.063), while the global test was not significant (p = 0.27). Exploratory effect-modification analyses did not identify heterogeneity of the time-varying lenva association by therapy line across pre-specified cutoffs (likelihood ratio tests for interaction p = 0.20 for 1 vs ≥2, p = 0.81 for 1–2 vs ≥3, and p = 0.13 for 1–3 vs ≥4), noting limited power and sparse events in early-line strata. Conclusions: LENVA-based salvage was not associated with improved OS vs CHEMO in aPD-1–refractory melanoma and effect estimates were time-dependent/uncertain amid strong confounding by therapy line. These data do not justify prioritizing a LENVA-vs-CHEMO prospective trial and highlight the need for new strategies in PD-1–resistant disease.

Reassessing obesity and smoking as renal cell carcinoma risk factors: TriNetX 2016-2025 analysis.

e16527 Background: Obesity and smoking are established risk factors for renal cell carcinoma (RCC). Most prior studies predate widespread adoption of ICD-10, which reliably differentiates between renal parenchymal and urothelial malignancies of the renal pelvis and ureter. Earlier studies grouped tumor types and used non-specific obesity definitions, potentially producing inaccurate risk estimates. A contemporary reassessment, employing precise tumor classification and detailed obesity definitions, is warranted. Methods: We conducted a retrospective cohort study using TriNetX. Patients (≥18 years) who received care using a general medical evaluation code (ICD-10 Z00) were included. Three exposure cohorts were defined: class I–II obesity (BMI 30.0–39.9 kg/m²), class III obesity (BMI ≥40.0 kg/m²), and smoking exposure. Patients with urothelial malignancies of the renal pelvis or ureter (C65–C66) were excluded. The primary outcome was incident RCC through 5 years. Propensity score matching (1:1) was performed. Obesity analyses were matched on clinical covariates, with smoking status included, while smoking analyses included BMI. Covariates included age, sex, race, ethnicity, hypertension, chronic kidney disease, type 2 diabetes mellitus, and chronic obstructive pulmonary disease. Analyses were conducted within the TriNetX platform. Patients were censored at the last follow-up. Results: The study included 2,487,294 patients with class I–II obesity, 1,779,240 with class III obesity, and 2,396,453 smokers. RCC incidence was 0.44% in class I–II obesity, 0.52% in class III, and 0.43% among smokers, compared with 0.20–0.23% in controls. Obesity and smoking were independently associated with RCC risk. Class III obesity demonstrated the strongest association (relative risk [RR], 2.26; 95% CI, 2.21–2.31; P < 0.001), followed by class I–II obesity (RR, 1.94; 95% CI, 1.90–1.98; P = 0.001). Smoking demonstrated elevated RCC risk (RR, 1.57; 95% CI, 1.54–1.61; P < 0.001). Conclusions: In a large, contemporary U.S. study, obesity demonstrated a stronger and graded association with RCC risk than smoking. These findings suggest that earlier studies may have overestimated smoking-associated risk and underestimated obesity-associated risk due to the inclusion of urothelial malignancies and non-specific obesity coding. Smoking and obesity remain independent and modifiable risk factors for RCC. To our knowledge, this is the first large-scale study to reassess these risk factors specifically in renal parenchymal tumors, excluding renal pelvis tumors. Exposure Group Patients (n) RCC Events (n) Relative Risk Ratio for RCC (95% CI) P Value Obesity Class I–II (BMI 30-39.9) 2487294 10879 1.94 (1.90–1.98) 0.001 Class III Obesity (BMI 40) 1779240 9310 2.26 (2.21–2.31) <0.001 Non-obese Controls 15382621 30355 Reference Smokers 2396453 10318 1.57 (1.54–1.61) <0.001 Never Smokers 17811442 40804 Reference

Adult medulloblastoma: Long-term follow-up analysis of a single-center cohort.

e14104 Background: Medulloblastoma (MB) is a malignant embryonal neoplasm of the central nervous system (CNS) that is more common in children and rare in adults—a setting in which there is limited clinical evidence to guide treatment and prognostication. Report of an adult MB cohort treated exclusively at our institution focusing on therapies performed and oncological outcomes in the long term. Methods: This is an observational retrospective cohort study including patients aged 17 years or older with a histological diagnosis of MB. Survival curves were estimated using the Kaplan–Meier method and compared using the log-rank test. Hazard ratios (HRs) and corresponding 95% confidence intervals (CIs) were calculated using Cox proportional hazards regression models. Variables with clinical relevance were evaluated in univariable analyses to explore their association with OS and PFS. Level of significance was defined as a two-sided (alpha = 0.05). All statistical analyses were performed using R software (survival package). Results: From 2008 to 2025, 72 patients were included in the analysis. Median time of follow-up is 169,57 months. Forty-nine male and 23 female; median age at diagnosis 28 years. All patients underwent surgery; gross total resection in 52 patients (73,2%). Histological subtypes were nodular/desmoplastic found in 31 patients (43.1%); 55 patients (82,1%) were staged as Chang score M0. Sixty-one patients received any adjuvant therapy (84,7%), radiotherapy with concomitant vincristine in 35 patients (58.3%) and radiotherapy alone in 22 patients (37.2%). Forty-one patients (67.2%) received adjuvant chemotherapy using three-drug regimens. Recurrence was diagnosed in 27 patients (37,5%), with a median time from diagnosis to recurrence of 41 months, and 19 patients received salvage therapy. Median overall survival was 157 months for the whole cohort (95% CI: 98,9 – NA), and median progression-free survival 96 months (95% CI: 63 – NA). Variables associated with worse prognosis included absence of adjuvant therapy (HR 12.97; 95% CI: 5.44–30.91; p:7x10-9) and recurrence (HR 7.59; 95% CI: 3.29–17,54; p: 2×10−6). Variables correlated with improved OS included: concomitant vincristine and radiotherapy (HR 0,4; 95% CI: 0,16 - 1; p: 0,05), and use of adjuvant chemotherapy (HR 0,3; 95% CI: 0,14-0,63; p: 0,001). Gross total resection (GTR) was not associated with improved OS (HR 0.76; 95% CI: 0.34–1.7; p: 0,5). Conclusions: This work represents the largest single-center cohort reported of adult medulloblastoma. This study highlights the importance of employment of adjuvant therapy following initial surgery and underlines the favorable prognosis in the adult MB patients.

Faith in the balance: Examining religious and spiritual influences on cancer treatment decisions.

e23268 Background: Religion and spirituality are fundamental aspects of many individuals’ lives, providing guidance, comfort, and strength during times of illness. Previous research has shown that religious beliefs can significantly impact cancer patients’ treatment decisions. The role of religion and spirituality in patient decision-making, especially in oncology settings, is complex and multifaceted, often creating ethical dilemmas that healthcare providers must navigate carefully. Significant gaps remain in understanding how these beliefs interact with medical recommendations, particularly in life-threatening situations. Less is known about how oncology providers manage these conflicts. This qualitative, cross-sectional observational study explores how religious and spiritual beliefs influence cancer patients’ treatment decisions and how oncology providers navigate this terrain. Methods: Cancer patients and oncologists were invited to participate in one-on-one semi-structured interviews. Patients were asked open-ended questions about their views on the role of religion and/or spirituality in their decision-making, their willingness to accept or decline specific treatments, and the role of divine intervention or miracles in their care. The Trust in Medical Research Scale (TMR) was also included. Oncologists were asked to share their experiences when religious and/or spiritual beliefs conflicted with medical recommendations and their strategies to ensure appropriate care. All interviews were audio-recorded with participant consent, transcribed, and deidentified for analysis. Results: Forty-eight patients and thirteen providers were interviewed. 77% of patients identified as religious or spiritual, with 67% claiming their beliefs influenced their medical decision making. Only one patient reported experiencing a conflict between their beliefs and medical recommendations. When asked to describe a time when a patient’s beliefs significantly impacted their treatment decisions, 38% of providers recalled patients refusing treatment/chemotherapy and 54% recalled Jehovah’s Witnesses declining blood transfusions. 70% of providers expressed feeling prepared when these circumstances arise. When asked how providers should respond when religious beliefs conflict with medical advice, 16% of patients favored nonjudgmental discussion, 27% suggested advising the patient that the best treatment was the science supported treatment, and 57% believed providers should simply respect the patient’s decision. Conclusions: Religious and spiritual beliefs can shape cancer patients' decision-making, and providers must be prepared to provide informed, autonomous patient care.