Citations

Abstract INTRODUCTION: Cholangiocarcinoma (CCA) is a deadly cancer of the hepatic bile duct epithelial cells. Loss-of-function mutations of epigenetic regulator ubiquitin C-terminal hydrolase deubiquitinase BRCA1-associated protein 1 (BAP1) occur in 26-32% of human CCA and are associated with worse prognosis and resistance to cytotoxic therapy. The goal of this study is to identify how BAP1 loss promotes cell death evasion through metabolic rewiring. MATERIALS & METHODS: Isogenic wild-type and BAP1-deficient human and murine CCA cell lines were generated with shRNA/siRNA or CRISPR/Cas9 and were used to assess ferroptosis sensitivity by treatment with ferroptosis inducers. In vitro, ferroptosis was assessed by phospholipid peroxidation (C-11BODIPY), mitochondrial phospholipid peroxidation (MitoPeDPP, MitoCLox), iron oeverload (MitoFerro green) and cell death (CellTiter-Glo). Phospholipidomics and proteomics were performed in isogenic cells by LC-MS/MS. Isogenic liver orthotopic CCA murine models were used for in vivo pre-clinical studies. RESULTS: Our findings indicate that BAP1 loss confers CCA ferroptosis resistance. BAP1 loss protects CCA against mitochondrial iron overload, ROS accumulation, and phospholipid peroxidation. Strikingly, BAP1 loss protects from mitochondrial phospholipid peroxidation and cardiolipin oxidation. Unbiased phospholipidomic profiling revealed that BAP1 loss remodels membrane phospholipid composition, enriching for ferroptosis-blocking monounsaturated phospholipids (MUFA-PLs) at the expense of ferroptosis-inducing polyunsaturated (PUFA-PLs). We found that BAP1 loss upregulates TLCD1, a phosphatidylethanolamine acyltransferase that incorporates MUFAs into membrane phospholipids, thereby stabilizing membranes against peroxidation. TLCD1 knockdown in BAP1-deficient CCA cells, restored ferroptosis sensitivity Notably, TLCD1 is significantly upregulated in human BAP1-mutant CCA as compared to WT tumors. We identified TLCD1 in the endoplasmic reticulum, Golgi and by mitochondrial outer membrane. In syngeneic liver orthotopic CCA murine models, engineered EpCAM-aptamer-coated lactosomes carrying GPX4 siRNA selectively targeted CCA cells, inducing ferroptosis and robustly suppressing tumor growth without systemic toxicity. Lastly, BAP1-mutant patient-derived xenografts (PDXs) and organoids were resistant to ferroptosis inducers, as compared to WT controls. CONCLUSIONS: Our findings reveal that BAP1 loss rewires mitochondrial phospholipid metabolism thus conferring CCA resistance to ferroptosis, identifying TLCD1 as a targetable metabolic vulnerability in BAP1-mutant CCA. Citation Format: Yu Zhao, Peyton Classon, Danielle Marie Carlson, Jayla Millender, Irene K. Yan, Sumera Ilyas, Rory L. Smoot, Gregory J. Gores, TUSHAR PATEL, Davide Povero. BAP1 loss confers ferroptosis resistance to cholangiocarcinoma via TLCD1-mediated membrane phospholipid remodeling [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 2433.

Abstract 5-Fluorouracil (5-FU) is a widely prescribed chemotherapy for colorectal, breast, gastric, and head and neck cancers. The drug inhibits thymidylate synthase, blocking DNA synthesis and causing cytotoxic stress that halts tumor growth. However, patients carrying deleterious variants in DPYD, the gene encoding the rate-limiting enzyme in 5-FU metabolism (dihydropyrimidine dehydrogenase, DPD), can experience fatal toxicity due to impaired drug clearance and accumulation of 5-FU metabolites. Current pharmacogenetic screening guidelines in the U.S. recommend testing for few of the >2,000 nonsynonymous variants that have been reported for DPYD, leaving patients with rare, uncharacterized mutations at risk. To interpret expanded testing, however, a means to classify DPYD variants of unknown significance is needed. To address this, we developed MPACT-DPD, a random-forest-based classifier that accurately predicts the functional impact of DPYD missense variants. Our model was trained on in-vitro activity of 156 variants and leveraged a feature set of biochemical, evolutionary, and AlphaFold3-derived structural features. We optimized hyperparameters using ten-fold stratified cross-validation and evaluated model performance with Matthews correlation coefficient (MCC) to account for moderate class imbalance (7:3 neutral to deleterious). It achieved exceptional performance, with a Matthews correlation coeffient (MCC) of 0.906 and an accuracy of 95.1% on an independent validation set (n=43). Furthermore, a SHAP (SHapley Additive exPlanations)-based interpretability analysis revealed cofactor proximity and residue conservation as the key drivers of predictions. MPACT-DPD showed superior performance at variant classification against other tools, including a DPYD gene-specific variant classifier, and has the potential to expand pre-treatment genetic screening to improve the safety of personalized 5-FU-based chemotherapy. Citation Format: Lulu Jiang, Ryan Jonathan Swartz, Lauryn Allyn Hahn, Brianna Bembenek, Hannah Marie Krause, Kelly Bouchonville, Steven M. Offer. MPACT-DPD: An interpretable machine learning classifier for predicting the functional impact of DPYD missense variants [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 2712.

Abstract Introduction: Src Family Kinases (SFKs) have a crucial role in tumor proliferation, survival, and metastasis. NXP900 is a highly selective SFK inhibitor with a novel mechanism of action that locks SFKs in a closed, inhibited conformation, providing sustained suppression of catalytic and non-catalytic functions, which is distinct from other SFK inhibitors like dasatinib. A phase I clinical trial of NXP900 in patients with advanced solid tumors is ongoing. Our preclinical studies showed therapeutic efficacy of NXP900 in cholangiocarcinoma (CCA) through decreased YAP, RAF-MAP and PI3K-AKT signaling. Multiomic profiling (RNA-seq, proteomics, phosphoproteomics) of CCA patient-derived xenografts implicated AKT pathway activation in primary NXP900 resistance. NXP900-resistant CCA cell lines generated via prolonged sublethal dose escalation showed a strong IL13RA2-AKT signature on multiomic analysis. Thus, we hypothesized that IL13RA2-AKT signaling was a mechanism of NXP900 resistance. We aimed to define this mechanism and demonstrate that the combinatorial targeting of IL13RA2 or AKT resensitizes to NXP900 treatment. Methods: HuCCT1 or HuCCT1-Resisant (HuCCT1-R) cells were injected bilaterally into the flanks of NOD/SCID mice and were allowed to grow for 3 months prior to tumor collection and analysis. Tumor engraftment and growth between HuCCT1 and HuCCT1-R groups were compared. To delineate the mechanisms for acquired NXP900 resistance, in-vitro treatment studies were performed using siRNAs targeting IL13RA2 as well as combination treatment with capivasertib, a pan-AKT inhibitor. Results: The HuCCT1-R cell line exhibited a higher rate of tumor engraftment and formed significantly larger tumors compared to the parental HuCCT1 cell line, indicating enhanced tumorigenic potential associated with acquired resistance. Silencing of IL13RA2 in HuCCT1-R cells resulted in pronounced reduction of cell proliferation and partially restored sensitivity to NXP900 treatment. Immunoblot analysis demonstrated a corresponding reduction in phosphorylated AKT (Thr308), consistent with suppression of downstream PI3K-AKT signaling. Furthermore, combination treatment of HuCCT1-R cells with the AKT inhibitor capivasertib, and NXP900 led to a significant increase in cell death and a greater inhibition in cell proliferation compared to either agent alone, indicating enhanced responsiveness and improved therapeutic efficacy. Conclusion: These findings confirm the central role of IL13RA2 signaling and the PI3K-AKT pathway in mediating acquired resistance to NXP900, consistent with our previous multiomics analysis. Importantly, this study identifies combination strategies that may be able to overcome NXP900 resistance, supporting the development of combinatorial approaches to restore therapeutic efficacy in resistant tumors. Citation Format: Shelby K. Yee, Hendrien Kuipers, Danielle M. Carlson, Jack W. Sample, Enis Ozmert, Hidemi Nishi, Erik Jessen, Dong-Gi Mun, Aushinie M. Abeynayake, Jennifer L. Tomlinson, Amro M. Abdelrahman, Nathan W. Werneburg, Binbin Li, Mitesh J. Borad, Mark J. Truty, Sumera I. Ilyas, Gregory J. Gores, Rory L. Smoot. Resistance to the SFK inhibitor NXP900 in cholangiocarcinoma is characterized by IL13RA2-AKT signaling and can be overcome by combination therapy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 7124.

Abstract Background: Cholangiocarcinoma (CCA) is a highly lethal epithelial cell malignancy of the hepatic biliary tract. Efficacy of current standard-of-care therapy is limited. Ferroptosis, an iron-catalyzed, caspase-independent cell death driven by phospholipid peroxidation, has emerged as a novel therapeutic target. Our goal is to develop a tumor-selective therapeutic approach targeting ferroptosis in CCA using siRNA-based nanotherapeutics. Materials & Methods: Milk-derived nanovesicles (lactosomes) coated with EpCAM aptamer and packaged with siRNA against Glutathione Peroxidase 4 (siGPX4-lactosomes) and Ferroptosis Suppressor Protein 1 (siFSP1-lactosomes), were used to induce ferroptosis in murine CCA cells (FAC) and in syngeneic liver orthotopic CCA-bearing mice. In vitro, lactosome uptake and efficacy were assessed by fluorescence-based, cell-viability and molecular biology assays. In vivo, target engagement, tissue distribution, safety and efficacy were assessed. Results: EpCAM (epithelial cell adhesion molecule expressed on CCA cells)-coated lactosomes showed robust uptake by FAC cells in a EpCAM-dependent manner. EpCAM knockdown via siRNA reduced lactosome uptake, confirming aptamer selectivity. Dose-dependent siGPX4/siFSP1-lactosome treatment led to GPX4/FSP1 knockdown and induced ferroptotic cell death in up to 60% of FAC cells after 72 hours (p<0.001 vs. siNC). In CCA-bearing mice, intravenous administration of siRNA lactosomes was well-tolerated with no liver or systemic toxicity observed and demonstrated robust tumor distribution compared to non-tumorous tissues. Efficacy studies demonstrated that siGPX4/siFSP1-lactosomes significantly reduced intratumor Gpx4 and FSP1 mRNA expression by 40-60% and induced significant upregulation of ferroptosis markers Ptgs2 and Acsl4 (p<0.01 vs. siNC). Ferroptosis in siRNA-lactosome treated tumors was confirmed by immunohistochemistry for phospholipid peroxidation markers oxPAPC and 4-HNE. Additionally, proliferation marker Ki67 was significantly downregulated in siGPX4/siFSP1-lactosome treated tumors (p<0.05 vs. siNC), indicating ferroptosis-mediated tumor suppression. Importantly, siGPX4/siFSP1-lactosomes reduced tumor burden and expression of CCA markers CK19/7 in treated mice, as compared to control mice. Conclusions: siRNA-based nanotherapeutics targeting CCA can effectively and safely induce ferroptosis in vitro and in vivo, representing a potential therapeutic strategy for CCA treatment. Citation Format: Peyton Classon, Danielle Marie Carlson, Jayla Millender, Irene Yan, Sumera Ilyas, Rory L. Smoot, TUSHAR PATEL, Gregory J. Gores, Davide Povero. Targeting cholangiocarcinoma with ferroptosis-inducing siRNA-based nanotherapeutics [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 5664.

Background Oral anti-cancer treatment is commonly used, and non-adherence can negatively affect outcomes. The Oral Oncolytic Non-Adherence Estimator (ORACLE) was designed to prospectively categorize patients into risk levels by considering constructs such as self-efficacy, health literacy, perception of treatment, quality of life, social support, and regimen complexity. Validating this estimator is essential. Methods A content validation survey evaluated representativeness, clarity, and factor structure of each construct in ORACLE. A multidisciplinary panel of 8 content experts and 4 lay experts completed the survey. Results were analyzed to evaluate inter-rater agreement, content validity index (CVI), and factorial validity index (FVI). Post-survey workgroup discussions reviewed results and considered tool adjustments. Results All 12 invited experts completed the survey. The average CVI for content experts was 0.88 for representativeness and 0.98 for clarity. The average CVI for lay experts was 0.95 for clarity. The average FVI was 0.88, with self-efficacy and negative perception constructs yielding the lowest scores (0.63 and 0.75, respectively). ORACLE revisions addressed survey results and expert feedback on health literacy, social support, and regimen complexity. Conclusions Strong CVI for content and lay experts (94% and 95%) and strong FVI for content experts (98%) verify that the tool is understandable and addresses non-adherence issues, establishing validity of construct clarity and representativeness. Content validity of the first pre-treatment oral oncolytic non-adherence risk estimator was established, providing a reliable tool to identify patients at higher risk of non-adherence.

PURPOSE Chronic myelomonocytic leukemia (CMML) is a rare myeloid neoplasm characterized by clinical heterogeneity and is associated with poor outcomes. To date, limited molecular information has been incorporated into disease classification and risk stratification. We aimed to integrate genomic features into the clinical decision-making process for CMML. PATIENTS AND METHODS We analyzed a retrospective cohort of 3013 patients with CMML (training set) and a prospective population of 516 patients (validation set). Using an innovative framework for multimodal data analysis, we developed molecular-based disease taxonomy and prognostication. RESULTS Unsupervised clustering identified nine entities with distinct genomic features and outcomes ( P < .001), including splicing machinery, transcription factors, signal transduction and tyrosine kinase pathways aberrations, and high-risk molecular signatures. Notably, 15% of patients showed molecular/clinical overlap with other myeloid neoplasms. We integrated molecular and clinical information to build the international CMML Prognostic Scoring System (iCPSS), incorporating mutations in nine genes together with hematologic parameters and cytogenetic abnormalities. The iCPSS identified five groups with distinct probability of overall and leukemia-free survival in both training and validation cohorts ( P < .001), outperforming existing prognostic models. Importantly, 55% of patients were reassigned to higher or lower risk groups by the iCPSS. Decision analysis demonstrated that iCPSS could refine the optimal timing of allogeneic transplantation at the individual level; compared with conventional prognostic tools, iCPSS-based decision modeling changed transplantation strategy in 31% of cases, resulting in a significant gain-in-life expectancy for eligible patient population ( P < .001). A federated learning platform was implemented to enable continuous, privacy-preserving model update across multiple centers. CONCLUSION Molecular information improves CMML classification and prognostication, supports more effective clinical decision making, and potentially refines the design of clinical trials.

Abstract SARS-CoV-2 infection disrupts iron homeostasis in organs such as the lung. Emerging evidence suggests similar dysregulation in the placenta may impair function and fetal growth. Ferroptosis, an iron-dependent form of cell death, is driven by the accumulation of labile ferrous iron, catalyzing the peroxidation of polyunsaturated fatty acids in the absence of sufficient iron storage or antioxidant defenses. This pathway has been implicated in SARS-CoV-2-related tissue damage in other organs and may contribute to placental dysfunction during pregnancy. However, the influence of infection timing, symptom severity, and fetal sex on placental iron regulation remains unexamined. We analyzed placental samples from a cohort of SARS-CoV-2-exposed pregnancies in the first (n = 6), second (n = 14), or third trimester (n = 30), classified by symptom severity (asymptomatic n = 23, symptomatic n = 27) and fetal sex, alongside unexposed controls (n = 47). We assessed placental iron deposition, lipid peroxidation, and mRNA and protein expression of ferroptosis markers. Publicly available RNA-sequencing datasets from human placenta tissue, exposed or not to SARS-CoV-2, were analyzed to assess ferroptosis-related gene expression. SARS-CoV-2 infection, particularly in symptomatic cases, was associated with reduced placental weight and birth weight. Asymptomatic cases showed increased placental iron deposition, which correlated with lower birthweight and was accompanied by elevated expression of nuclear receptor coactivator 4, a cytosolic adaptor protein that mediates the selective autophagic degradation of ferritin (ferritinophagy) and iron release. Placentas from symptomatic patients exhibited evidence of altered iron transport and sex-specific down-regulation of antioxidant defenses. Transcriptomic analyses further suggested widespread disruption of ferroptosis pathways in placentas from infected patients. Our findings reveal that SARS-CoV-2 infection alters placental iron homeostasis and is associated with ferroptosis-related changes, with distinct molecular responses based on timing of infection, symptom severity, and fetal sex.

Abstract Background Progressive supranuclear palsy (PSP) is a tauopathy for which there is limited understanding of epidemiological risk factors. Objective The objective of this study was to identify premorbid nonmedical and medical risk factors for PSP diagnosis in a large prospective population‐based cohort. Methods We performed a matched nested case–control study using the UK Biobank (UKB) cohort consisting of >500,000 adults aged 37 to 73 years. PSP diagnoses were identified from 2007 to censor date (August 2025) and matched 1:10 by age and sex to control subjects without parkinsonism or dementia diagnoses. We derived risk factors using an iterative logistic regression approach. Results There were 240 incident PSP diagnoses in the UKB. In the fully adjusted model, premorbid depression (odds ratio [OR], 3.22; 95% confidence interval [CI], 2.30–4.50; P < 0.001), delirium (OR, 6.76; 95% CI, 4.08–11.19; P < 0.001), and functional gastrointestinal disorders (OR, 1.91; 95% CI, 1.39–2.63; P < 0.001) were associated with increased risk. Heavy alcohol consumption was associated with higher risk compared with moderate consumption (OR, 1.65; 95% CI: 1.21–2.26; P = 0.0017). Body mass index (BMI) and alcohol consumption had nonlinear risk profiles. Cancer diagnosis was inversely associated with PSP (OR, 0.56; 95% CI, 0.40–0.79; P < 0.001). Depression and delirium diagnoses remained significant 5 to 10 years before diagnosis. Conclusions In this large prospective cohort, PSP was associated with premorbid lifestyle practices and neuropsychiatric diagnoses years before diagnosis. This suggests a long premotor phase and highlights opportunities for earlier diagnosis and further mechanistic investigation. The inverse association with cancer mirrors the relationship seen in other neurodegenerative conditions and may point toward shared mechanisms. © 2026 International Parkinson and Movement Disorder Society.

Abstract Background Artificial intelligence (AI)–based, mammography breast cancer risk prediction models show improved discriminatory accuracy relative to clinical risk models. However, data on their calibration are limited. This study compared model performance of 3 clinical breast cancer risk models: Gail, Tyrer–Cuzick v8, and Breast Cancer Surveillance Consortium v3 with the MIRAI AI-risk model. Methods Digital mammograms were ascertained from a screening mammography cohort of 12 308 women within the Mayo Clinic Biobank with 250 incident breast cancers (176 invasive) within 5 years. We predicted 5-year breast cancer risk, estimated discriminatory accuracy (C-index), and calibration (observed-to-expected ratio) of both overall and invasive breast cancer and compared estimates using bootstrapping approaches. Results MIRAI demonstrated similar or improved discriminatory accuracy of overall breast cancer (C-index = 0.71, 95% confidence interval [CI] = 0.68 to 0.74) and invasive breast cancer (C-index = 0.71, 95% CI = 0.67 to 0.75) compared with clinical models (overall breast cancer: C-index = 0.59-0.68; invasive breast cancer: C-index = 0.60-0.68). MIRAI’s calibration for risk of overall breast cancer (observed-to-expected ratio = 0.96, 95% CI = 0.85 to 1.08) was improved compared with Gail (observed-to-expected ratio = 1.22, 95% CI = 1.07 to 1.38) and Breast Cancer Surveillance Consortium (observed-to-expected ratio = 1.38, 95% CI = 1.22 to 1.56) but similar to TC with volumetric percent density and polygenic risk score (observed-to-expected ratio = 0.99, 95% CI = 0.87 to 1.13). However, for low-risk women (approximately 50%), MIRAI overestimated risk of overall breast cancer. MIRAI also overestimated risk of invasive breast cancer across the risk spectrum (observed-to-expected ratio = 0.68, 95% CI = 0.58 to 0.78), whereas clinical models had good calibration (observed-to-expected ratio = 0.86–0.99). Conclusion MIRAI demonstrated stronger discriminatory accuracy than clinical models for 5-year overall and invasive breast cancer risk prediction but overestimated risk for both breast cancer endpoints. AI-based risk models should consider discriminatory accuracy and calibration for invasive cancer before implementation.