Browse Articles

A phase 2 multi-center pharmacodynamic study of the fatty acid synthase (FASN) inhibitor TVB-2640 in advanced <i>KRAS</i> -mutant non–small cell lung cancer.

3079 Background: Preclinical and translational studies demonstrate that KRAS-mutant NSCLC has a unique metabolic dependence on de novo fatty acid synthesis, mediated by fatty acid synthase (FASN) and downstream lipid metabolism pathways that support saturated fatty acid production, enabling rapid tumor growth, survival, and resistance to lipid peroxidation. TVB-2640 is an orally bioavailable and selective FASN inhibitor that blocks the β-ketoacyl reductase step of fatty acid synthesis and has demonstrated antitumor activity in KRAS mutant NSCLC models. Methods: This single-arm trial (NCT03808558) enrolled patients with previously treated, advanced KRAS mutant NSCLC. Eligible patients had progressed on prior platinum-based chemotherapy (100%) and immune checkpoint inhibition (95%), had measurable disease by RECIST v1.1, ECOG performance status 0–1, and adequate organ function. One patient had received prior KRAS directed therapy. TVB-2640 100 mg/m 2 (equivalent to two to four 50-mg tablets) was administered orally once daily in continuous 28-day cycles. Exploratory correlative analyses included 11 C-acetate PET, plasma lipidomics, sebaceous lipid profiling, and quality-of-life measures. Results: Among 18 enrolled patients, the median age was 69 years, 12 (67%) were male, 14 (78%) were white and 13 (72%) had a history of smoking. Median number of prior lines of therapy was 4. The most common KRAS mutations were G12V (44%), G12C (17%), and G12D (17%). No patients had a radiographic response, 7 (39%) had stable disease, 9 (50%) had disease progression, and 2 (11%) were non-evaluable. Five patients (28%) had radiographic shrinkage of target lesions but did not achieve partial response due to the magnitude of decrease (n = 3) or progression of non-target lesions (n = 2). Median progression-free survival was 1.8 months (95% CI, 1.7-1.8 months). Overall, 84% of patients had adverse events (AE). Most common AEs were xerostomia (50%), dry skin (44%), and fatigue (39%). Most common grade ≥ 3 AEs (44%) were fatigue, dermatological toxicities, xerostomia, ocular toxicities and anemia (11% each). Conclusions: Single-agent FASN inhibition with TVB-2640 in heavily pretreated KRAS mutant NSCLC did not meet the primary endpoint of response. Since FASN is a downstream effector of mutant KRAS, combination regimens with KRAS inhibitors can be considered as a future approach in NSCLC. Ongoing correlative studies may further inform the use of this agent. Despite not meeting its primary endpoint, this clinical trial highlights a novel therapy targeting energy metabolism in lung cancer. Clinical trial information: NCT03808558 .

PTPRZ1–MET fusion and induction of reprogrammed immuno-microenvironment via MET–IFN-β–LGALS9 signaling pathway during IDH-mutant glioma progression.

e14088 Background: Recurrent IDH-mutant high-grade gliomas are difficult to treat, with limited effective targeted or immune therapies. PTPRZ1–MET (ZM) fusion is a recurrent alteration enriched during malignant progression, present in ~15% of secondary glioblastomas and associated with significantly worse survival (median OS 127 vs 248 days in ZM-negative sGBM). However, how ZM fusion reshapes the immune microenvironment in recurrent IDH-mutant gliomas and whether it creates targetable immune vulnerabilities remain unclear. Here, we investigated ZM fusion–associated immune remodeling and evaluated the rationale for combined MET and LGALS9 targeting. Methods: Single-cell RNA sequencing was performed on six recurrent IDH-mutant high-grade gliomas. Myeloid populations were subclustered and annotated using module-based approaches. Ligand–receptor analyses compared intercellular communication by fusion status. A fusion-associated macrophage signature was used to build prognostic models via Cox regression, LASSO, and random forest selection, followed by Kaplan–Meier analysis. MET and LGALS9 expression were validated by immunohistochemistry and multiplex immunofluorescence with spatial line-scan analysis. ZM fusion was introduced into glioma cell models and confirmed by Western blot and imaging. Therapeutic efficacy was evaluated in an immunocompetent orthotopic IDH-mutant/ZM fusion GL261 model treated with anti-LGALS9 antibody (RG9-1), the MET inhibitor vebreltinib, or their combination. Results: scRNA-seq revealed marked myeloid heterogeneity and identified a ZM fusion–associated microglia-like macrophage program characterized by enhanced antigen presentation and myeloid activation, which supported a prognostic risk score stratifying patient outcomes. Cell–cell communication analysis showed signaling rewiring in fusion-positive tumors, highlighting strengthened MET–LGALS9 interactions between tumor and myeloid compartments. Transcriptomic profiling demonstrated elevated type I interferon responses in fusion-positive tumor cells, with altered IFN-I regulatory nodes in both tumor and microglial populations. Tissue validation confirmed cell-type specificity, with higher MET expression in tumor cells and higher LGALS9 in myeloid/microglial cells (P&lt;0.001), as well as close spatial proximity. In vivo, combined RG9-1 and vebreltinib treatment achieved the strongest tumor suppression and longest survival compared with monotherapy or control. Conclusions: PTPRZ1–MET fusion defines a recurrent subset of IDH-mutant gliomas with immunosuppressive remodeling. Integrated evidence supports a fusion-associated MET–IFN-β–LGALS9 axis linking tumor-intrinsic interferon signaling to macrophage-mediated immune regulation, providing a rationale for combined MET and LGALS9 targeting.

National trends and disparities in cancer-related mortality among adults with diabetes in the U.S., 1999–2023.

e21003 Background: The prevalence of diabetes in the U.S. has nearly doubled, contributing to increased cancer risk due to inflammation and obesity, etc. While cancer incidence of adults with diabetes has been studied, overall national mortality trends remain sporadic. Methods: Cancer-related mortality data from 1999 to 2023 from the CDC WONDER database was analyzed retrospectively. Data included U.S. adults aged ≥ 65 years, ICD-10: E10 – E14 (diabetes) and ICD-10: C00 – D49 (neoplasm-related mortality). Age-adjusted mortality rates (AAMR) per 100,000 were calculated using the 2000 U.S. Standard Population – with annual percentage change (APC) determined regarding mortality trends. Results: Overall, there were 778,936 deaths from cancer among adults with diabetes (AAMR: 69.8). The total AAMR increased from 1999–2005 (+1.7%/year), and 2018–2021 (+10.3%/year), and the total AAMR decreased from 2005–2018 (–1.2%/year) and 2021–2023 (–2.6%/year). Men and Non-Hispanic American Indian/Alaskan Native individuals had the highest AAMRs. Non-Hispanic White and Hispanic individuals had slight increases till 2021 followed by declines. Among census regions, the South and Midwest had higher AAMRs. There were 1999-2007 and 2018-2020 increases in metro areas, while non-metropolitan areas showed rises in 1999–2003 and 2018–2020. Conclusions: Improvements in cancer care have not reduced cancer-related mortality for adults with diabetes. We observe discrepancies by race, age, region, etc. We need targeted prevention and equitable care for high-risk populations to decrease cancer mortality.

Causes and trends of under-five mortality in the Gilgel Gibe Health and Demographic Surveillance System, 2005 to 2019, Southwest Ethiopia: Cohort study

Background Irrespective of intensive global efforts to reduce under-five mortality, it remains a significant public health concern. Understanding the causes and trends of under-five mortality is essential for guiding targeted interventions, assessing the effectiveness of public health strategies, and monitoring changes in mortality over time. The health and demographic surveillance system is one of the preferred sources to study the cause of under-five mortality. Objective This study aims to analyse the causes and trends of under-five deaths in Southwest Ethiopia using Gilgel-Gibe Health and Demographic Surveillance System (GGHDSS) database. Methods GGHDSS is an open, dynamic cohort that was established in 2005. Fifteen years of mortality data and seven years of Verbal Autopsy (VA) data were extracted from the GGHDSS database for this study. The VA data are part of the fifteen year mortality dataset, in which causes of death were determined through the VA method. After extracting the data from MYSQL and OpenHDS, it was exported to Excel for further cleaning. Finally, the cleaned data were exported to R statistical software for analysis and visualization. Neonatal, infant, and under-five mortality trends were analysed and the proportion of cause specific deaths were identified from the VA data. Results Between 2005 and 2019, 28, 811 children, 13931 (48.35%) female and 14880 (51.65%) male, were born alive and registered in the GGHDSS, among which 1828 (6.34%) of them died before celebrating their fifth birthday. The overall under-five mortality rate was 63.4 (95% CIs; 60.6, 66.3) per 1000 live births while neonatal and infant mortality rates were 30(95% CIs; 28.1, 32) and 50.2 (95% CIs; 47.6, 52.7), respectively. The mortality rate during the surveillance years showed a declining trend among neonatal, infant, and overall under-five children, with average slopes of −2.49, −4.56, and −7.24, respectively. From 1070 under-five deaths captured by the VA method in GGHDSS during 2009–2016, 596 (55.7%) were male and 474 (44.30%) were female. The most common causes of neonatal death were birth asphyxia and perinatal respiratory disorders, bacterial sepsis, and prematurity including respiratory distress. In the post-neonatal period, the most common causes of death were acute lower respiratory infections (including pneumonia and acute bronchitis), intestinal infectious diseases including diarrheal diseases, and malaria. Severe malnutrition, intestinal infectious diseases, and acute lower respiratory infections were responsible for more than half of the deaths among children between 12 and 59 months of age. Conclusions Under-five mortality has shown a significant declining trend between 2005 and 2019 at the study setting. Birth asphyxia, neonatal infections, and prematurity were among the most common causes of neonatal deaths, whereas infectious diseases and malnutrition were the main causes of death beyond the neonatal period. Strengthening perinatal and neonatal care, improving prevention and management of childhood infections, and enhancing early nutritional interventions are needed to reduce under-five mortality.

Emerging nanotechnology trends in skin and hair care cosmeceuticals

A qualitative study to reveal devastating impacts of chronic fracture-related infection

Insights into the role of internal catalysts in asparagine-to-succinimide conversion in a hyperthermophilic glutamine amidotransferase

Phase I study of the LiViT regimen (liposomal irinotecan, vincristine, and temozolomide) in pediatric patients with relapsed/refractory solid tumors.

10039 Background: Relapsed or refractory solid tumors in children and adolescents remain a major therapeutic challenge with limited effective options. Liposomal irinotecan provides prolonged systemic exposure compared to conventional irinotecan and may offer improved efficacy and tolerability. We conducted a phase I trial to determine the recommended phase II dose (RP2D), safety, and preliminary antitumor activity of the LiViT regimen in patients aged 2–18 years. Methods: This single-center, open-label phase I study enrolled patients aged 2–18 years with relapsed/refractory solid tumors. The LiViT regimen consisted of vincristine (1.5 mg/m² IV day 1; max 2 mg), temozolomide (150 mg/m²/day IV days 1–5) and liposomal irinotecan at escalating doses of 40, 50, 65, 80, or 100 mg/m² (IV day 1) every 3 weeks. A standard 3+3 dose-escalation design was used (Ia), followed by an expansion cohort (Ib). Results: Twenty-nine patients were enrolled (Ia: 18; Ib: 11). Diagnoses included rhabdomyosarcoma (n=12), neuroblastoma (n=9), desmoplastic small round cell tumor (n=3), Ewing sarcoma (n=2), and others (n=3). In phase Ia, liposomal irinotecan was administered at escalating doses of 40, 50, 65, 80, and 100 mg/m². At 80 mg/m², 1 of the first 3 patients experienced a dose-limiting toxicity (DLT) (grade 3 diarrhea lasting &gt;3 days despite loperamide); the cohort was expanded to 6 patients, and no additional DLTs or other grade 3–4 toxicities were observed. No DLTs occurred among patients treated at 100 mg/m² during Ia (n=3) or in the Ib expansion cohort (n=11). The MTD was not reached, and 100 mg/m² was selected as the RP2D based on safety in 14 patients. Across all dose levels, the only grade ≥3 treatment-related adverse event was the single episode of grade 3 diarrhea at 80 mg/m²; all other toxicities were grade 1–2. Among 27 evaluable patients, 5 achieved partial response (PR) and 12 had stable disease (SD), yielding an objective response rate (ORR) of 18.5% and a disease control rate (DCR) of 63.0%. Conclusions: The LiViT regimen is feasible with a manageable safety profile in pediatric patients with relapsed/refractory solid tumors. The MTD was not reached, and liposomal irinotecan at 100 mg/m²—the highest tested dose—was established as the RP2D. Promising preliminary activity supports phase II evaluation. Clinical trial information: NCT06710821 .

First-in-human PET imaging of ⁶⁸ Ga-FC516, a novel ACP3-targeted radiotracer for prostate cancer.

5128 Background: Prostatic acid phosphatase (ACP3) is overexpressed in more than 85% of prostate cancers, and largely absent in normal tissues, resulting in over 100-fold tumor-to-background specificity. ACP3 represents a differentiated imaging biomarker that may complement prostate-specific membrane antigen (PSMA), particularly in PSMA-low or PSMA-treated patients. We developed 68 Ga-FC516, a novel ACP3-targeted radiotracer, and evaluated its preclinical characteristics and first-in-human imaging performance. Methods: 68 Ga-FC516 was studied preclinically using in vitro assays (radiochemical stability, enzyme inhibition, cellular uptake, internalization, and efflux) and in vivo PET/CT studies, including blocking and biodistribution experiments. A first-in-human exploratory clinical trial was conducted in healthy volunteers and prostate cancer patients. The primary objective was to assess safety and radiation dosimetry; secondary objectives included biodistribution assessment and lesion detection. PET/CT imaging results were compared with PSMA radiotracers or 68 Ga-OncoACP3-DOTA when possible. Results: 68 Ga-FC516 was labeled efficiently with radiochemical purity exceeding 95%. In in vitro studies, 68 Ga-FC516 demonstrated potent target binding, long cellular retention and high stability. ACP3-specific tumor uptake of 68 Ga-FC516 was confirmed via an in vivo blocking experiment. The tracer showed low background uptake in liver, kidneys, and salivary glands, with tumor uptake exceeding 16.4% ID/g in a biodistribution study. Gallbladder uptake was markedly lower than that of 68 Ga-OncoACP3-DOTA (2.6% vs. 35.0% ID/g) at 1 hour post-injection. In an ongoing clinical trial in metastatic prostate cancer patients, increased uptake of 68 Ga-FC516 (SUV max in PET/CT) was observed at 1 to 3 hours across all lesions, with improved tumor-to-kidney ratios over time. The tracer also demonstrated low background uptake, including in salivary glands and kidneys, which was significantly lower than that of PSMA-targeted tracers. Notably, a patient with both prostate and rectal cancers showed strong uptake in the prostate tumor (SUV max 44), but no uptake in the rectal lesion, further establishing the tracer's superior specificity. No adverse events were observed. Conclusions: We have shown that 68 Ga-FC516 has high ACP3-targeting specificity, low background uptake and excellent imaging performance in preclinical and first-in-human clinical studies. These findings support the potential of ACP3 as a differentiated biomarker for prostate cancer diagnosis and patient stratification, complementing PSMA PET and addressing unmet needs in PSMA-low or PSMA-treated patient populations. The clinical validation study was approved by the Ethics Committee of the First Affiliated Hospital of Jinan University (KY-2025-254). Research Sponsor: FindCure Biosciences Co., Ltd. Clinical trial information: KY-2025-254.

IFCT-2401 SPORADIC trial: Integrating cemiplimab to sequential hypofractionated chemoradiotherapy for unfit or elderly patients with unresectable stage III non–small cell lung cancer.

TPS8128 Background: Patients with unresectable stage III non-small cell lung cancer (NSCLC) who are elderly or have significant comorbidities are frequently ineligible for concurrent chemoradiotherapy (CCRT). Sequential chemoradiotherapy (seq-CRT) remains the preferred approach in this population. While consolidation immunotherapy after CCRT improves survival, evidence supporting its optimal integration with seq-CRT is limited. Preclinical and clinical data suggest that combining chemotherapy and immunotherapy enhances tumor immunogenicity through increased antigen release, immune priming, and modulation of the tumor microenvironment. Hypofractionated radiotherapy may further potentiate immune-mediated effects while shortening overall treatment duration, which is particularly relevant for frail patients. The IFCT-2401 SPORADIC trial evaluates in unfit or elderly patients a sequential strategy integrating neoadjuvant chemo-immunotherapy prior to hypofractionated radiotherapy, followed by maintenance immunotherapy, in patients with unresectable stage III NSCLC not eligible for CCRT. Exploratory correlative studies include circulating biomarkers, immune profiling, lymphopenia, and metabolic response assessment. Methods: IFCT-2401 SPORADIC is a multicenter, randomized, open-label phase II trial (EU-CT 2024-517316-29-00). Patients are randomized in a 2:1 ratio to receive neoadjuvant chemotherapy alone (arm A) or neoadjuvant chemo-immunotherapy (arm B). Neoadjuvant treatment consists of three 28-day cycles of carboplatin (AUC 5, day 1) and paclitaxel (80 mg/m², days 1, 8 and 15). Cemiplimab (350 mg, every 3 weeks) is added in arm B. All patients subsequently receive curative hypofractionated thoracic radiotherapy (55 Gy in 20 fractions), followed by maintenance cemiplimab every 3 weeks for up to 12 months. Eligible patients have unresectable stage IIIA–C NSCLC. Three categories of patients are considered unfit for CCRT: age ≥70 years with ECOG performance status (PS) 0 to 1, age &lt; 70 years with ECOG PS 0 to 1 and significant comorbidities or age &lt; 70 years with ECOG PS 2. Key exclusion criteria include actionable oncogenic drivers, prior systemic therapy for NSCLC and active autoimmune disease. The primary endpoint is progression-free survival. A total of 152 patients are planned for enrollment across 25 centers. First patient was enrolled on November 17, 2025. Clinical trial information: EU-CT 2024-517316-29-00.

Extent of resection and survival in <i>IDH</i> -mutant WHO grade 2 glioma: A systematic review and meta-analysis.

e14098 Background: For IDH-mutant WHO grade 2 gliomas, extent of resection (EOR) is thought to improve long-term outcomes, but optimal thresholds and survival benefit remain uncertain. Quantifying the impact of EOR can guide surgical planning and patient counseling. Methods: We performed a systematic review and meta-analysis of MEDLINE, Embase, CENTRAL, and Web of Science. Eligible studies reported EOR (continuous or categorical) and outcomes including overall survival (OS), progression-free survival (PFS), malignant transformation, seizure control, and treatment-related morbidity. Random-effects models pooled hazard ratios (HRs) for survival outcomes, and meta-regression assessed relationships between EOR, histologic subtype (astrocytoma vs oligodendroglioma), measurement method (MRI volumetrics vs surgeon estimate), and adjuvant therapy. Results: Twelve studies including 1,052 patients were analyzed. Higher EOR was associated with improved survival: pooled HR for OS per 10% incremental resection was 0.88 (95% CI 0.83–0.93), and for PFS 0.85 (95% CI 0.79–0.91). Patients achieving ≥90% resection had the greatest benefit, with 10-year OS of 88% versus 62% for &lt;90% resection (p &lt; 0.001). Higher EOR also reduced risk of malignant transformation (HR 0.67, 95% CI 0.53–0.85) and improved seizure control in 68% of evaluable patients. Results were consistent across histologic subtypes and measurement methods, with minimal heterogeneity (I² &lt; 35%). Conclusions: Greater EOR in IDH-mutant grade 2 gliomas is strongly associated with improved survival, lower malignant transformation risk, and better seizure control, supporting maximal safe resection as the standard of care. Clinical takeaway: Surgical planning aiming for ≥90% resection can optimize long-term outcomes and quality of life in patients with IDH-mutant grade 2 gliomas.

Digital health intervention for comorbid insomnia and neurocognitive impairment in adult survivors of childhood cancer: Results from a randomized clinical trial.

10003 Background: Survivors of childhood cancer are at risk of developing both insomnia and cognitive impairment, which may persist decades following treatment. Cognitive behavioral therapy for insomnia (CBT-I) is the gold-standard treatment for insomnia, yet the impact of improved sleep on comorbid cognitive dysfunction has not been evaluated in survivors. Methods: Survivors enrolled in the Childhood Cancer Survivor Study (CCSS; median age 44 [range 23-66] years; 73% female; 37% leukemia, 23% lymphoma, 18% sarcoma, 14% Wilms, 8% neuroblastoma) with comorbid insomnia and neurocognitive impairment were randomized 1:1 to 9-week digital CBT-I vs. online sleep education (SE). Insomnia severity index (ISI), neurocognitive problems (CCSS-Neurocognitive Questionnaire [NCQ]), emotional health (Patient Health Questionnaire/Generalized Anxiety Disorder), and quality of life (PROMIS) were measured at baseline and 9-week and 6-month follow-ups. Intent-to-treat models estimated treatment effects for ISI (mean difference [MD] between CBT-I and SE of change from baseline to follow-ups) adjusted for age, sex, and cranial radiation. Cohen’s d estimated effect size. Additional models estimated treatment effects on secondary outcomes and mediation by improved ISI. Results: 439 of 541 randomized survivors completed baseline assessments and started the intervention (82% CBT-I; 81% SE); 353 of those completed 9-week (76% CBT-I; 85% SE) and 305 completed 6-month (61% CBT-I; 78% SE) post-intervention assessments. Survivors randomized to CBT-I reported improved ISI scores at 9 weeks (MD=-2.4[95%CI -3.4, -1.4]; Cohen’s d =0.50) and 6 months post-intervention (MD=- 3.5[-4.6, -2.4]; d =0.72) vs. SE (both p’s&lt;0.001). Survivors randomized to CBT-I reported greater improvement in memory (9wks: MD=-0.3SD [-0.5, -0.1], p=0.003; 6mos: MD=-0.5SD [-0.7, -0.3], p&lt;0.001) and task efficiency (9wks: MD=-0.3SD [-0.5, -0.1], p=0.003; 6mos: MD=-0.5SD [-0.7, -0.3], p&lt;0.001) vs. SE. Effects of CBT-I on cognitive function were mediated by improved ISI scores (42-65% mediated, all p’s &lt;0.001). Insomnia responders (those who reported a change in ISI from &gt;8 [mild to severe insomnia] to &lt;8 [no insomnia]) also reported improved quality of life (9wks: mental beta=1.7[95%CI 0.4, 3.0], physical=2.1 [1.0, 3.1]; 6mos: mental=3.5 [2.0, 5.0], physical=3.3 [2.1, 4.5]), and emotional health (9wks: anxiety beta=-1.3 [-2.1,-0.5], depression=-1.8[-2.6, -1.0]; 6mos: anxiety=-1.8 [-2.7, -0.8], depression=-2.4 [-3.4, -1.4]). Conclusions: Use of a fully automated digital health treatment of insomnia confers immediate and long-term benefits to comorbid cognitive impairment and quality of life in survivors of childhood cancer. Given its efficacy and availability, digital CBT-I should be considered for the behavioral treatment of insomnia and comorbid symptoms in survivors. Clinical trial information: NCT04317742 .

KIF11 as a druggable target in NSCLC: An integrated computational analysis.

e20716 Background: Non-small cell lung cancer (NSCLC) in non-smoking women represents a biologically distinct subgroup in which actionable therapeutic targets can be identified and may also have broader relevance across diverse NSCLC subgroups. This study aimed to identify hub genes from tumor-normal transcriptomics and to identify druggable targets using drug–gene interaction mining, structure-based docking, and in silico developability profiling. Methods: Microarray data from the publicly accessible dataset GSE19804 (Affymetrix U133 Plus 2.0), comprising 60 tumors and 60 paired adjacent normal lung tissues, were analyzed utilizing GEO2R/limma. Differentially expressed genes were identified with a significance threshold of p ≤ 0.05 and |logFC| ≥ 1.0. Protein–protein interactions (PPI) were constructed using STRING with a combined score of ≥ 0.700 and a false discovery rate (FDR) of 1%, and these interactions were ranked by degree centrality in Cytoscape/cytoHubba. Results: PPI results revealed ten highly connected upregulated hubs: CDK1, CCNB1, CCNA2, KIF11, BUB1, TOP2A, DLGAP5, NUF2, MELK, and PBK, with a high-confidence and robust integration of KIF11 (degree 57; mean score 0.973). DGIdb analysis prioritized KIF11 as the most actionable hub, led by ispinesib–KIF11 and filanesib, followed by litronesib, AZD-4877, dimethylenastron, S-Trityl-L-cystein (STLC). Docking to the KIF11 L5 allosteric pocket (PDB 4A5Y) showed stable binding across inhibitors (–8.7 to –10.5 kcal/mol), with ispinesib and STLC strongest (–10.5), driven by Glu116-centered hydrogen bonding and extensive hydrophobic/aromatic contacts. OSIRIS and ADMET predictions indicated overall oral drug-like properties, high intestinal absorption, and no predicted AMES mutagenicity, while differentiating safety/liability profiles (Table). Notably, STLC and dimethylenastron exhibited more favorable predicted safety profiles, including lower hERG and hepatotoxicity probabilities, compared with ispinesib, which showed higher liability signals. Conclusions: This study highlights STLC as a promising KIF11 inhibitor with strong structural engagement and favorable in silico developability, alongside ispinesib as a validated reference, supporting KIF11 as a central, druggable vulnerability in female non-smoking NSCLC with potential relevance across broader NSCLC subgroups. In silico binding, ADMET, and OSIRIS drug-likeness profiles of KIF11 inhibitors. Drug Binding Affinity (kcal/mol) HIA AMES hERG Risk Overall Drug Score (OSIRIS) Ispinesib –10.5 High Negative High 0.36 S-Trityl-L-cysteine (STLC) –10.5 High Negative Low 0.38 Filanesib –9.0 High Negative Low 0.35 AZD-4877 –9.1 High Negative Moderate 0.56 Dimethylenastron –9.0 High Negative Very Low 0.56 Litronesib –8.7 High Negative Low 0.33 Abbreviations: HIA, human intestinal absorption; AMES, Ames mutagenicity test; hERG, human ether-a-go-go–related gene channel.

Efficacy and safety of neoadjuvant chemotherapy with AG regimen with or without nimotuzumab in borderline resectable or locally advanced pancreatic cancer: A retrospective cohort study.

e16375 Background: This study aimed to evaluate whether adding nimotuzumab to an albumin-bound paclitaxel plus gemcitabine (AG) regimen could improve outcomes in these patients. Methods: We analyzed data from 100 patients with BRPC/LAPC. Patients were divided into a control group (AG alone, n=57) and a study group (AG+nimotuzumab, n=43). The primary endpoint was overall survival (OS). Results: The disease control rate was 100% in both groups. The objective response rate was higher in the study group (12% vs. 4%, P=0.238), accompanied by a significantly greater mean reduction in tumor diameter (-13.7% vs. -8.2%, P=0.019). The R0 resection rate was comparable (81.4% vs. 87.7%, P=0.552). As of May 5, 2025, the median OS (mOS) was not significantly different between groups (25 vs. 24 months, P=0.229). However, in the high EGFR expression subgroup (n=50), the study group showed significantly longer mOS (33 vs. 23 months, P=0.016). Similarly, in this subgroup, the study group had favorable trends in median PFS (23 vs. 16 months, P=0.051) and significantly longer median disease-free survival (25 vs. 16 months, P=0.0025). The median PFS in the overall population was 18 vs. 17 months (P=0.279). The safety profile was comparable, with grade 3 adverse events occurring in 7.0% (control) and 4.7% (study) of patients. Conclusions: The addition of nimotuzumab to neoadjuvant AG chemotherapy demonstrated promising efficacy, particularly in patients with high EGFR expression, and was well-tolerated. These results support further investigation in prospective randomized trials.

Prevalence of <i>CDH1</i> germline variants in gastric cancer patients with exome-inferred ancestry.

4073 Background: Germline pathogenic or likely pathogenic (P/LP) CDH1 variants increase the lifetime risk of developing diffuse gastric cancer (DGC). Given that the epidemiology and prognosis of gastric cancer vary based on patient ancestry, and Hispanic and Black/African-American patients have been underrepresented in clinical and translational research, we sought to determine the prevalence of CDH1 variants in a diverse population of gastric cancer patients. Methods: Germline whole exome sequencing (WES) data from gastric cancer patients were obtained as part of commercial ordering of personalized circulating tumor DNA testing (Signatera, Natera, Inc.) from March 2024 to June 2025. Ancestry was inferred by EthSeq and classified into predefined groups: European (EUR), admixed American (AMR), African (AFR), and East Asian (EAS). Germline single-nucleotide variants and short insertions/deletions were identified and annotated with Variant Effect Predictor (v.105), and the pathogenicity of variants in CDH1 was determined using classifications available in the ClinVar database. The prevalence of P/LP variants and variants of uncertain significance (VUS) were compared across ancestries. Potential pathogenicity of VUS was evaluated using in silico predictors SIFT (v5.2.2), Polyphen (v2.2.2), and AlphaMissense (hg19). Results: Among 2,567 patients, 55% (1,414/2,567) were male and the median age was 67 years. Ancestry inference identified 47% (n=1,201) as EUR, 22% (n=575) as AMR, 18% (n=457) as AFR, and 13% (n=334) as EAS. Patients with AMR ancestry were younger compared to EUR ancestry patients (median 62 years vs 69 years; adj p&lt;0.0001), and less likely to be male (51% [295/575] vs. 58% [693/1201]; OR, 0.7; 95% CI, 0.63–0.94; p=0.015). No significant differences in age or sex were observed in patients with AFR (median 68 years, 55% male) or EAS (median 68 years, 52% male) ancestry compared to EUR. Overall, 0.7% (n=17) gastric cancer patients had a P/LP variant, and there was no difference across ancestries (0.8% of EUR, 0.5% of AMR, 0.4% of AFR, and 0.6% of EAS patients). We found that 1.1% (n=27) of the cohort had a CDH1 VUS, the rate for which did not vary across ancestries (1.2% of EUR, 0.52% of AMR, 0.84% of AFR, 0.60 of EAS). Of the 21 unique VUS, 33% (n=7) were each predicted to be pathogenic by all three in silico predictors. One unique VUS was observed in two unrelated patients: NM_004360.5:c.635G&gt;T (p.Gly212Val). The predicted-pathogenic VUS were similarly distributed across ancestries. Conclusions: This real-world analysis highlights that CDH1 VUS were similarly distributed across ancestries. Notably, we found that one-third of VUS identified were predicted to have pathogenic function across three in silico prediction tools. This highlights the need for functional assays to determine the variant biology to improve gastric cancer risk assessment.

Testicular plasmacytoma: A comprehensive case series and review.

e19565 Background: Testicular plasmacytoma (TP) is a rare manifestation of plasma cell neoplasms, occurring alone or with multiple myeloma (MM). Diagnosis is challenging due to nonspecific presentation, and literature is limited to small case reports. We aimed to characterize TP’s clinical presentation, management, and outcomes at a large tertiary center. Methods: We retrospectively reviewed over 7,000 PCN patients treated between 2007 and 2025. Patients with TP confirmed by imaging, biopsy, or orchiectomy and complete data were included. A concurrent literature review was performed. Demographics, labs, pathology, imaging, treatments, and survival outcomes were analyzed. Results: A literature review identified 34 cases and 5 clinical patterns of TP based on disease timing and progression. Our case series included 17 TP patients, the largest single-institution cohort to date. Eighty-eight percent (patterns 2, 3, and 5) had underlying MM, with type 4 most common. Median age was 57 years, and IgG was the predominant isotype. Over half had osseous plasmacytomas on PET/CT, although most lacked diffuse bone marrow involvement at diagnosis. Median overall survival was ~4 years, with one patient surviving beyond 18 years. Conclusions: TP is extremely rare and usually represents secondary involvement in active or relapsed MM rather than de novo disease. No cases of primary TP were observed, highlighting the importance of long-term follow-up to detect progression to systemic disease. Radical inguinal orchiectomy remains central for diagnosis and local control, with radiation used selectively. Systemic therapy follows standard myeloma regimens, and prognosis is primarily determined by underlying MM biology. Literature review: Patient characteristics and five clinical patterns. Patient Cohort No. of patients Age, years Presentation M-Protein MM Status Treatment De novo solitary TP (Pattern 1) 11 72 (34–86) Painless scrotal swelling; unilateral (9); bilateral metachronous (2) Absent No prior MM or extramedullary PCN Orchiectomy ± XRT (1) or CT (2) TP as the first manifestation of MM (Pattern 2) 3 69 (62–73) Painless scrotal swelling without systemic symptoms Absent (2); IgA-λ light chain restricted (1) New MM diagnosis Orchiectomy + CT ± XRT (1) TP during active multiple myeloma (Pattern 3) 3 47 (43-72) Painless scrotal swelling λ light chain restricted (3) Active, relapsed or residual Orchiectomy ± CT (2) TP with other extramedullary PCN (Pattern 4) 7 57 (38–77) Painless scrotal swelling Absent (2), IgA (1), IgG (1), or λ light chain restricted (3) Active (3), relapsed (1) or no disease (3) Orchiectomy ± CT (5) ± ASCT (3) TP as MM relapse (Pattern 5) 10 54 (39–81) Painless or painful scrotal swelling IgA (5) or IgG (2), λ light chain restricted (7) Relapsed MM Orchiectomy, CT ± XRT (2) ± ASCT (1) LN: lymph nodes, PCN: plasma cell neoplasm; MM: Multiple Myeloma; CT: chemotherapy; XRT: radiation; ASCT: autologous stem cell transplant.

Bridging the cardio-oncology care gap in India: Real-world guideline adherence and AI-driven workflow optimization in 2300 breast cancer patients at a tertiary cardiac center.

e12719 Background: Cardiotoxic cancer therapies necessitate guideline-recommended cardiac surveillance in breast cancer patients, yet real-world implementation and efficiency remain variable. Data from low- and middle-income settings are limited. We evaluated real-world adherence to ASCO cardio-oncology guidelines and the impact of AI-assisted echocardiography on monitoring efficiency and clinical response in a large breast cancer cohort. Methods: We conducted a retrospective real-world analysis of 2,300 breast cancer patients receiving potentially cardiotoxic therapy at a tertiary center. Guideline adherence was assessed using ASCO recommendations for baseline and surveillance echocardiography. AI-assisted echocardiographic LVEF measurements were compared with standard reporting. Primary outcomes included timeliness of cardiac imaging and reporting, detection of cancer therapy–related cardiac dysfunction, and subsequent clinical management. Results: A total of 2,300 breast cancer patients receiving potentially cardiotoxic systemic therapy were included (mean age 52 ± 11 years). Anthracycline-based regimens were administered in 1,426 patients (62%), and anti-HER2 therapy in 644 (28%). Cardiovascular comorbidities were common, including hypertension (38%), type 2 diabetes mellitus (31%), and obesity (24%). Baseline adherence to guideline-recommended echocardiography was 94% (95% CI, 93–95%). Prior to AI implementation, adherence to serial surveillance was 72% (95% CI, 70–74%), which significantly increased to 89% (95% CI, 88–91%) following AI-enabled workflow integration. Mean echocardiographic analysis time per study decreased by 74% (2.1 vs 8.2 minutes; p &lt; 0.001). Cancer therapy–related cardiac dysfunction (CTRCD) occurred in 11.8% of patients (n = 271; 95% CI, 10.5–13.2%). All patients with CTRCD received guideline-directed cardioprotective therapy. During follow-up, mean LVEF improved by 7.25% (95% CI, 6.4–8.1%), and 95% of patients remained free from heart failure hospitalization. On multivariable analysis, diabetes mellitus (OR 2.8; 95% CI, 1.9–4.1) and age &gt; 60 years (OR 2.1; 95% CI, 1.4–3.2) independently predicted CTRCD. AI-derived LVEF demonstrated strong agreement with expert assessment (accuracy 85% [95% CI, 83–87%]; precision 81%; recall 73%; F1 score 67%). Conclusions: Using ASCO recommendations for baseline and surveillance echocardiography, this real-world study identified a significant cardio-oncology care gap that was substantially reduced through AI-enabled workflow integration. Improved guideline adherence enabled timely detection and management of cardiotoxicity, resulting in meaningful LVEF recovery and low heart failure hospitalization rates in a resource-constrained setting.

A multiple-instance learning-based deep learning model with multi-channel input for early detection of breast tumor.

e12576 Background: Early detection of breast cancer is essential but remains limited by the high cost. Echo signal–based approaches offer a low-cost alternative and have demonstrated tumor detection using single-channel signals; however, limited information content restricts performance stability. Recent advances in multi-channel echo signal acquisition highlight the need for adaptive models that can effectively integrate different channel configurations. Methods: Single-channel echo signals were prospectively collected from 200 breast tumor patients who underwent curative surgery at Seoul National University Hospital (SNUH) and SMG-SNU Boramae Medical Center between Oct 2022 and Feb 2024. In addition, four-channel echo signal data were collected from 34 patients between Apr and Dec 2025 at SNUH. To enable unified processing of different channel configurations, all inputs were represented in a bag-of-instances format, with single-channel data modeled as a single-instance bag and four-channel data as a four-instance bag. Feature extraction was performed using an ImageNet-pretrained EfficientNet-B0 backbone, followed by a gated attention–based multiple instance learning architecture. Results: In the independent test set, single-channel models achieved AUC values ranging from 0.739 to 0.872, whereas the proposed model achieved an AUC of 0.892. Although the difference in AUC was not statistically significant (DeLong test, p = 0.7804), the AUC standard deviation of the four-channel–based MIL model (0.007) was substantially lower than that of the single-channel models (0.015–0.043), indicating superior performance stability. Notably, the MIL model showed improved performance in the independent test set compared with internal evaluation, suggesting enhanced generalization. Conclusions: This study demonstrates that extending single-channel echo signal–based approaches to a four-channel–based MIL framework improves performance stability and generalization in breast tumor detection. Although the performance difference was not statistically significant, the proposed method demonstrates the potential of integrating multiple channels for more robust and stable predictions in clinical applications. Model performance comparison in terms of AUC and others. Model AUC ACC@SEN95 SPE@SEN95 PPV@SEN95 NPV@SEN95 1 st CH Single 0.872±0.025 0.813±0.025 0.815±0.052 0.802±0.045 0.825±0.009 2 nd CH Single 0.865±0.034 0.807±0.029 0.810±0.055 0.795±0.046 0.819±0.010 3 rd CH Single 0.823±0.043 0.751±0.068 0.696±0.139 0.717±0.086 0.798±0.025 4 th CH Single 0.739±0.015 0.673±0.037 0.536±0.064 0.619±0.035 0.768±0.033 4CH-MIL 0.892±0.007 0.841±0.058 0.863±0.092 0.850±0.095 0.837±0.030 Abbreviations: AUC, Area under the curve; ACC, Accuracy; SEN, Sensivitiy; SPE, Specificity; PPV, Positive predictive value; NPV, Negative predictive value.

Optimizing denosumab treatment strategies to balance skeletal-related events for prevention and osteonecrosis of the jaw risk in metastatic breast cancer patients: A Korean nationwide target trial emulation.

1103 Background: Administration of bone-modifying agents (BMAs) is essential for reducing the incidence of skeletal-related events (SREs) in breast cancer bone metastases patients. As denosumab showed superior effect for prevention of SREs compared to zoledronic acid, denosumab is commonly used BMA in this setting; however, the optimal administration interval and treatment duration of denosumab remain unclear. Methods: This study utilized the Health Insurance Review and Assessment Service (HIRA) database and emulated a target trial using the clone–censor–weight approach. Among the overall breast cancer population, 2,023 metastatic breast cancer patients who received regular denosumab administration were identified during Jan 2009 to Aug 2023. We examined the administration patterns and duration of denosumab therapy and evaluated their associations with the incidence of skeletal-related events (SREs) and osteonecrosis of the jaw (ONJ). Results: During follow-up, approximately 23% of patients continued regular denosumab administration at intervals of less than 8 weeks. Patients who received denosumab for ≥25 months exhibited a low risk of skeletal-related events (SREs) (3.42–6.15%, according to treatment compliance); however, the cumulative risk of osteonecrosis of the jaw (ONJ) increased to 19.47% (95% confidence interval [CI], 13.79–26.75) at 4 years of follow-up. Among patients treated with denosumab for 13–24 months, the risk of SREs was comparable to that observed in those treated for ≥25 months (5.15–6.77% vs. 6.15%), while the risk of ONJ remained relatively low (10.02–10.19%, depending on compliance). In contrast, patients who received denosumab for less than 12 months demonstrated a substantially higher risk of SREs, reaching 19.44% (95% CI, 6.47–45.69). Shorter administration intervals were associated with increased ONJ risk but did not significantly affect SRE prevention. Conclusions: This nationwide real-world study is the largest to date to evaluate the optimal duration and administration interval of denosumab in metastatic breast cancer patients with bone metastases. To effectively prevent SREs while minimizing the risk of ONJ, a denosumab treatment duration of 13–24 months appears to offer the most favorable benefit-risk balance and may be considered an optimal treatment strategy in clinical practice.

Impact of AI-guided multi-analyte precision oncology–based platform on clinical outcomes in treatment-refractory pan-cancer patients.

e15121 Background: Refractory solid tumors exhibit profound molecular and functional heterogeneity, frequently requiring repeated treatment modifications after failure of standard therapies. Single-parameter testing is insufficient to guide such iterative decision-making. Comprehensive, integrated tumor profiling may enable rational, personalized treatment iterations and sustain clinical benefit across successive lines of therapy. Methods: Patients with refractory solid tumors underwent integrated EXACTA profiling encompassing genomic alterations, gene expression, immunohistochemistry, immune biomarkers, and in-vitro chemosensitivity assays. Actionable alterations and targetable pathways were identified to guide personalized therapy across successive lines. Progression-free survival (PFS) was assessed for each EXACTA-guided line. PFS ratios (PFSr) were calculated relative to the immediately preceding line, with modified PFSr (mPFSr) applied to adjust for potential false-positive and false-negative signals. The best mPFSr per patient was used to assess clinical benefit from iterative EXACTA-guided therapy. Results: EXACTA based profiling identified a highly heterogeneous yet clinically actionable genomic landscape, with recurrent alterations in PIK3CA , EGFR , ERBB2 , KRAS , BRAF , CCND1 , and BRCA2 supporting precision-guided therapy. PD-L1-stratified analysis revealed enrichment of truncating APC mutations in PD-L1-positive tumors, whereas CTNNB1 alterations were confined to PD-L1-negative tumors. Sequential EXACTA-guided therapies derived from a single comprehensive profiling event resulted in clinically meaningful benefit across multiple treatment lines. Based on best modified progression-free survival ratios (mPFSr) achieved per patient, 60.74% of patients demonstrated mPFSr &gt; 1.5, 68.14% achieved mPFSr &gt; 1.3, and 68.8% achieved mPFSr &gt; 1.25. Line-wise distributions of mPFSr, including the proportion of patients achieving mPFSr &gt; 1.3 at each EXACTA-guided treatment line, are summarized in the Table 1. Conclusions: Integrated EXACTA profiling revealed distinct tumor biology and enabled optimized treatment strategies. Mutual exclusivity of CTNNB1 alterations in PD-L1–negative tumors and enrichment of truncating APC mutations support β-catenin–TCF4–mediated PD-L1 regulation. In refractory solid tumors, a single EXACTA profiling enabled iterative, biology-guided therapy with sustained benefit. PFS ratio gains beyond the first EXACTA-guided line demonstrate the durable clinical utility of integrated profiling, supporting improved outcomes despite disease progression. Durability of Clinical Benefit Across EXACTA-Guided Lines. mPFSr from Exacta Lines &gt;1.5 &gt;1.3 &gt;1.25 L2/L1 30.43% 39.13% 39.13% L3/L2 22.22% 25.92% 22.22% L4/L3 50% 50% 50%