Citations

Abstract Purpose: This multicenter, multi-surgeon retrospective study aimed to identify risk factors for postoperative cerebrospinal fluid (CSF) leak following endonasal resection of craniopharyngioma, while evaluating the "perception gap" through a surgeon survey. Methods: A retrospective review was conducted on 416 patients who underwent endoscopic transsphenoidal surgery (ETS) for craniopharyngioma from 20 institutions. Factors were compared between patients with and without postoperative CSF leak, and between Early (2007-2015) and Late (2016-2025) Epochs. Complementing the clinical data, a survey of 19 neurosurgeons captured expert perspectives on risk stratification and management strategies. Results: Overall postoperative CSF leak rate was 13.5% (56/416 patients). Univariate analysis identified predominantly cystic tumors (34% vs. 21%, p = 0.034)and intraoperative lumbar drain (LD) use ( p < 0.028) as associated with postoperative CSF leak. BMI ( p = 0.587), prior surgery ( p = 0.576), and tumor size ( p = 0.363) were not significant. In the multivariable analysis, LD use was associated with a higher postoperative CSF leak rate (OR 1.91, 95% CI 1.06–3.46; p = 0.030). Between Epochs, nasoseptal flap (NSF) utilization increased from 70.7% to 87.6% ( p <0.001). NSF was protective (OR 0.28, 95% CI 0.07–0.92; p = 0.037) in the Early Epoch; no factors were significant in the Late Epoch. The surgeons’ survey identified prior surgery and intraoperative high flow leaks as primary risks; however, their LD protocols diverged from clinical data. Conclusion: Despite increased use of NSF, CSF leak persists. The study highlights a significant divergence between expert perception and clinical data regarding lumbar drains.

Abstract Clinical narrative text contains crucial patient information, yet reliable extraction remains challenging due to linguistic variability, documentation habits, and differences across care settings. Large language models (LLMs) have shown strong accuracy on clinical information extraction (IE), but their reproducibility (stability under repeated runs) and robustness (stability under small, natural prompt variations) are less consistently quantified, despite being central to clinical deployment. In this study, we evaluate three open-weight LLMs representing distinct modeling choices: a dense general-purpose model (Llama 3.3), a mixture-of-experts (MoE) general-purpose model (Llama 4), and a domain-tuned medical model (MedGemma). We focus on binary clinical IE aligned with four mobility classes from the International Classification of Functioning, Disability and Health (ICF) framework. Using a controlled experimental design, we quantify (1) intra-prompt reproducibility across repeated sampling and (2) inter-prompt robustness across paraphrased prompts. We jointly report predictive performance (F1-score) and stability (Fleiss’ Kappa). And we test factor effects using three-way ANOVA with post-hoc comparisons. Results show that increasing temperature generally degrades agreement, but the magnitude depends on model and task; furthermore, prompt paraphrasing can substantially reduce stability, with particularly large drops for the MoE model. Finally, we evaluate a practical mitigation, self-consistency via majority voting, which improves κ substantially and often improves or preserves F1-score, at the cost of additional inference. Together, these findings provide a reproducible framework and concrete recommendations for evaluating and improving LLM reliability in clinical IE.

Background Immune exclusion inhibits antitumor immunity and response to immunotherapy, but its mechanisms remain poorly defined. In triple-negative breast cancer (TNBC), an aggressive and generally immune-rich subtype, an immune-cold microenvironment predicts poor prognosis due to a limited response to chemotherapy and immune checkpoint inhibitors. This study aimed to identify mechanisms regulating immune infiltration in TNBC. Methods We performed spatial transcriptomic analysis comparing immune-enriched versus immune-cold treatment-naïve TNBCs. Functional analyses, including loss-of-function and reconstitution experiments, were conducted to investigate the role of trophoblast cell-surface antigen 2 (TROP2), a key target of anticancer antibody drug conjugates (ADCs), in promoting TNBC progression. A humanized TROP2 syngeneic TNBC model was used to assess the effects of TROP2-targeting in combination with anti-programmed cell death protein 1 (PD-1) therapy. Additionally, data from patients treated with immune checkpoint blockade were used to test hypotheses from the preclinical findings. Results We reveal that TROP2 controls barrier-mediated immune exclusion in TNBC through claudin 7 association and tight junction regulation. TROP2 expression is inversely correlated with T-cell infiltration and predicts poor outcomes in TNBC. We demonstrate that TROP2 is sufficient to drive tumor progression in vivo in a CD8 T cell-dependent manner, while its loss deregulates expression and localization of multiple tight junction proteins, enabling T-cell infiltration. We show that TROP2 targeting via hRS7, the antibody component of the ADC sacituzumab govitecan, enhances the anti-PD-1 response and improves T-cell accessibility and effector function. Correspondingly, TROP2 expression is highly associated with lack of response to anti-PD-1 therapy in human breast cancer. Conclusions This study defines a new mechanism of barrier-mediated immune exclusion in cancer controlled by TROP2-dependent tight junctions. This mechanism drives tumor progression but can be targeted via TROP2-directed therapy to activate antitumor immunity and enhance immunotherapy response.

Abstract Background: Triple-negative breast cancer (TNBC) is the most aggressive subtype of breast cancer and remains difficult to treat due to the absence of targeted therapies. While recent advances in immunotherapy, particularly with immune checkpoint inhibitors (ICIs), have transformed the treatment landscape for many cancers, response rates to ICIs remain low for TNBC, largely due to tumor-intrinsic immune evasion mechanisms. Objective: We identify and characterize a novel immune evasion pathway in TNBC driven by deregulation of cancer cell-intrinsic protease-activated receptor 1 (PAR1), a thrombin receptor frequently upregulated in aggressive tumors. Methods and Results: Through integrative bioinformatic analyses, genetic perturbations, xenograft models (immunodeficient and immunocompetent), flow cytometry-based tumor immune profiling, and RNA-seq, we demonstrate that PAR1 activation enhances immune evasion in TNBC. Mechanistically, we show that the paracaspase MALT1 acts as a critical downstream effector of PAR1 signaling and promotes immune escape by driving PD-L1 expression in TNBC cells. Functional assays reveal that genetic depletion of MALT1 or PD-L1 increases TNBC cell susceptibility to T cell-mediated cytotoxicity in vitro and significantly suppresses tumor growth in vivo. Notably, through knock-in and rescue experiments using a catalytically inactive MALT1 mutant, we reveal that MALT1’s scaffolding - rather than protease - function is essential for PD-L1 regulation. Immune cell depletion assays further identify CD4+ T cells, CD8+ T cells, and natural killer (NK) cells as key mediators of the anti-tumor immune response suppressed by the PAR1-MALT1 axis. These findings are supported by human TNBC specimen analyses, where MALT1 expression inversely correlates with T cell activation and positively associates with PD-L1 in PAR1-high, but not PAR1-low, tumors. Conclusion: Our study reveals immune evasion as a novel mechanism of PAR1-driven breast cancer pathogenesis, mediated by a PAR1/MALT1/PD-L1 signaling cascade. These findings bridge a critical gap by positioning MALT1 at the interface between tumor-intrinsic signaling and the immune microenvironment - two areas of CARMA-BCL10-MALT1 (CBM) complex research that have largely evolved in parallel over the years. Importantly, targeting MALT1 may enhance immunotherapy efficacy in TNBC. Given the availability of MALT1 inhibitors in clinical development, our work supports combining MALT1 inhibition with ICIs or other immune-oncology agents as a promising strategy to overcome immune resistance in TNBC. Citation Format: Dong Hu, Prasanna Ekambaram, Zheqi Li, Linda Klei, Maria L. Beecher, Yi Liu, Zongyou Cai, John Little, Jeffrey A. Meridew, Jia-Ying Lee, E. Aubrey Thompson, Tullia C. Bruno, Lidija Covic, Seung-Oe Lim, Anushka Dongre, Heide L. Ford, Mien-Chie Hung, Adrian V. Lee, Steffi Oesterreich, Linda McAllister-Lucas, Peter C. Lucas. A cancer cell-intrinsic PAR1/MALT1/PD-L1 signaling pathway drives immune evasion in triple-negative breast cancer [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 7392.

Abstract The INFORM trial (NCT01670500) was a randomized, two-arm Phase II neoadjuvant study comparing the efficacy of a platinum-based regimen (cisplatin) versus an anthracycline (AC)-based regimen in participants with germline BRCA1/2 mutations and early-stage, HER2-negative breast cancer. MicroRNAs (miRNAs) are emerging as promising non-invasive biomarkers for disease detection and treatment monitoring. The initial aim of this study was to evaluate whether pretreatment plasma miRNA profiles were associated with treatment outcomes in the INFORM trial cohort. Pretreatment plasma samples from 97 INFORM participants (n=53 cisplatin and n=44 AC) were screened for 352 miRNAs commonly implicated in cancer using the qPCR-based ID3EALTM Cancer Panel. Twenty out of 53 participants achieved residual cancer burden (RCB) score of 0 or 1 with cisplatin and 20/44 with AC. Higher plasma miR-362-3p expression was associated with a favorable response to cisplatin (1.7-fold; p&lt;0.01), but not to AC. MiR-362-3p expression in paired pretreatment tumor biopsies (n=79) did not significantly correlate with plasma expression and was not associated with RCB 0/1 (p&gt;0.05). In TCGA, miR-362-3p expression was higher in breast tumors than in adjacent normal tissue, and higher in triple negative breast cancers (TNBCs) compared with hormone receptor-positive tumors (both FDR&lt;0.001), independent of BRCA mutation or Single Base Substitution Signature 3 (SBS3) status. The miR-362-3p findings in INFORM did not replicate in a sister trial, TBCRC 030 (NCT01982448; 11 out of 46 cisplatin-treated patients achieved RCB 0/1, p=0.46). These results suggest that while miR-362-3p is unsuitable as a circulating biomarker for cisplatin response, it may play an important biological role in mediating cisplatin sensitivity in TNBCs, independent of BRCA1/2 mutations. Functional studies demonstrated that miR-362-3p is expressed and secreted by TNBC cell lines, operates as a tumor suppressor, and generally correlates with cisplatin sensitivity. Overexpression of miR-362-3p rendered MDA-MB-231 and CAL-51 cells cisplatin sensitive, while knockdown induced resistance in MDA-MB-436 cells. Ongoing studies are testing whether these effects translate in xenograft models. In silico and functional assays identified BCLAF1, a DNA damage response (DDR) regulator, as a direct target of miR-362-3p. Overexpression of miR-362-3p suppresses BCLAF1 mRNA and protein levels. Our findings demonstrate that miR-362-3p enhances cisplatin responsiveness by targeting BCLAF1. A miR-362-3p-based therapeutic may be useful as a co-agent to enhance tumor responses to platinum-based chemotherapy. Citation Format: Zhaoji Liu, Shizhong Ke, Xiaohui Li, Catherine Wu, Madison M. Uyemura, Brian R. Sardella, Erica S. Massicott, Lin Wang, Emily K. Aronson, Dimitra Karagkouni, Nikolas Kalavros, Ioannis S. Vlachos, Felipe Batalini, Cristina S. Bogsan, Jit Kong Cheong, Lihan Zhou, He Cheng, Phillip Munson, Erica L. Mayer, Judy E. Garber, Stuart J. Schnitt, Nadine M. Tung, Andrea L. Kasinski, Frank J. Slack, Gerburg M. Wulf, Yujing J. Heng. MiR-362-3p enhances platinum response in breast cancer [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 7903.

Abstract Radiation is a cornerstone of glioblastoma (GBM) treatment, however, the majority of tumors progress within the radiation field due to inherent radiotherapy (RT) resistance. The integration of highly potent radiosensitizers, such as the ATM inhibitors AZD1390 or WSD0628, could reverse resistance and significantly improve local control of these tumors. However, the achievable brain tumor exposure of these inhibitors may be limited by drug exposure in skin and mucosa, leading to enhanced radiation toxicity. To support the clinical development of WSD0628, we developed pharmacokinetic (PK)-efficacy models to help interpret ongoing PK analyses. A model based on in vitro cell studies was based on the in vitro observations that 30 nM WSD0628 provided maximum ATM inhibition and the extent of radiosensitization was directly related to the duration of drug incubation after irradiation, up to a maximum of 24 hours. The unbound WSD0628 concentration, measured by rapid equilibrium dialysis, was 66% of the total concentration in cell culture media, and was used to define an optimal radiosensitizing target of 20 nM free WSD0628 for 4 to 24 hours. Previous PK analyses showed that the free, unbound fraction (fu) of WSD0628 in plasma and brain tumor tissue are 0.02 and 0.054, respectively, and the brain tumor-to-plasma ratio is 0.32 in GBM43 patient-derived xenografts (PDXs). From these data, a total drug level of 1160 nM in plasma is predicted to achieve unbound drug levels above the 20 nM WSD0628 target unbound concentration. To validate this prediction, parallel in vivo dose-ranging studies were performed in orthotopic PDXs. Mice with orthotopic GBM43 were randomized to treatment with RT (8 Gy x1) alone or in combination with a range of WSD0628 doses. RT combined with 0.25 mg/kg and 1 mg/kg WSD0628 had no impact on survival, compared to RT alone, while median survival increased by 1.3-fold for RT combined with 2.5 mg/kg, 1.7-fold with 5 mg/kg, and 3.2-fold with 10 mg/kg WSD0628. A similar dose-response was observed in an M12 melanoma brain metastasis PDX. Based on PK modeling of these drug doses, 2.5 mg/kg dosing will maintain total WSD0628 plasma levels above 1160 nM for 11.9 hours, 5 mg/kg for 23.6 hours, and 10 mg/kg for 40 hours. Taken together, we have developed a model to predict efficacious dosing of a brain tumor based on the free drug hypothesis, which allows integration of multiple aspects of in vitro modeling with measured drug levels in animal models to predict total plasma levels associated with robust sensitizing effects. Assuming similar tissue to plasma partitioning in human GBM, this model could be directly applied to interpret the ongoing Phase 1 PK analysis of WSD0628 in recurrent GBM. Citation Format: Ann Catherine Mladek, Juhee Oh, Sneha Rathi, Lily Liu, Danielle M. Burgenske, Brett L. Carlson, Katrina K. Bakken, Lauren L. Ott, Zeng Hu, Wei Zhong, William F. Elmquist, Jann N. Sarkaria. Modeling pharmacokinetics and efficacy of the ATM inhibitor WSD0628 in GBM and melanoma metastasis PDX models [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 4629.

Abstract Prostate cancer requires continual escape from immune surveillance, but mechanisms underlying this escape in the transition from androgen sensitivity to resistance and metastasis remain incompletely understood. Androgen-deprivation therapy (ADT)—the treatment backbone—induces tumor cell apoptosis but also remodels the tumor microenvironment (TME). We have assembled a single-cell RNA-seq atlas from &gt;800,000 cells across 68 patients treated at Columbia University and Memorial Sloan Kettering. The resource includes 33 primary tumors (16 treatment-naïve, 17 post-ADT) annotated with time to PSA recurrence, and 31 metastases, including longitudinal pre- and post-ADT data in 14 castration-sensitive tumors (CSPC), 17 castration-resistant tumors (CRPC), and 4 neuroendocrine tumors (NEPC). This dataset spans the spectrum of prostate cancer disease progression.Using ARACNe and VIPER to infer activity of transcriptional regulatory and signaling proteins, we detect conserved cell state programs while minimizing batch effects. Several significant features emerged. Immune infiltration—both lymphoid and myeloid—is significantly higher in CSPC metastases than in primary or CRPC tumors and increases further after ADT. T cells constitute ∼15% of cells in primary tumors versus 30% in treatment-naive CSPC (40% post-ADT), falling to 13% in CRPC and &lt;1% in NEPC (p&lt;1e-32), consistent with an initially "hot" CSPC TME that progressively excludes T cells with disease progression.Among ∼350,000 epithelial/tumor cells, trajectory analysis reveals two differentiation pathways from normal epithelium. Pathway 1—dominant in primary tumors—shows androgen receptor (AR) activity with three sub-phenotypes (T1-T3): T1 exclusive to primary tumors; T2 shared by primary and CSPC; and T3 restricted to CRPC, with high ERG expression, and very strong AR activity. Pathway 2 (clusters T4-T6), present in NEPC and a subset of CRPC, lacks AR activity and displays SOX2, FOXA2, PRKD1. These aggressive states uniformly overexpress AURKA, RET, and TOP2A, suggesting druggable targets for combination therapy.We further define five cancer-associated fibroblast (CAF) subtypes. CAF1-3 track with advanced, treatment-resistant disease—enriched in NEPC versus CRPC (25% vs 18% of CAFs) and in CRPC versus CSPC (18% vs 11%)—while CAF4-5 predominate in primary tumors. We identify druggable proteins and receptor-ligand interactions linking CAF1-3 with T4-T6 tumor states; in vivo perturbations are underway.This atlas provides a resource profiling tumor and immune ecosystems across prostate cancer progression. It highlights MR programs and stromal-tumor-immune interactions that underlie immune evasion and ADT resistance, yielding actionable targets and combination-therapy hypotheses for aggressive disease. Citation Format: Aleksandar Obradovic, Kwangmin Yoo, Chang Liu, Patrick McCann, Casey R. Ager, Matthew C. Dallos, Samir Zaidi. Comprehensive profiling of primary and metastatic prostate tumors reveals distinct tumor and fibroblast cell states associated with androgen resistance [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 7419.

Abstract Background: Epidemiologic and other studies have identified environmental, occupational, and genetic risk factors for lymphoid and myeloid malignancies, but most studies have been conducted in Western populations. Investigations in populations with differing exposure patterns, distributions of disease subtypes, and genetic architecture are needed to fully understand hematopoietic tumor etiology. Methods: We conducted a large, multicenter, hospital-based case-control study of lymphoid and myeloid neoplasms in East Asia (AsiaLymph), including 5,671 lymphoid cases, 1,879 myeloid cases, and 3,858 controls. Participants completed a computer-assisted personal interview and provided biospecimens. Cases underwent central pathology review and were coded into the WHO Classification. Herein, we describe the study methods in detail and report association results for education, body mass index, and family history. Results: Greater BMI at age 20 but not at age 40 was associated with odds of total lymphoid neoplasm (per 5 kg/m2 increase: OR [95% CI]: 1.18 [1.09-1.27]), total myeloid neoplasm (OR [95% CI]: 1.17 [1.05-1.29]), and several subtypes. Greater educational attainment was associated with increased odds of total lymphoid neoplasm (for college vs. less than primary education: OR [95% CI]: 1.41 [1.21-1.65]) but not myeloid neoplasm (OR [95% CI]: 1.17 [0.95-1.45]; p-heterogeneity=0.02). There were also positive associations between family history of hematologic cancer in first degree relatives and odds of lymphoid neoplasm (OR [95% CI]: 1.53 [1.15-2.03]) and myeloid neoplasm (OR [95% CI]: 1.65 [1.11-2.44]) and specific subtypes. None of the evaluated risk factors were associated with NK/T-cell lymphoma, which supports a distinct etiology for this subtype. Conclusion: The AsiaLymph Study is one of the largest molecular epidemiology studies of both lymphoid and myeloid neoplasms with standardized World Health Organization classification of histopathologic subtypes and will serve as a valuable resource for etiologic investigations into risk factors for these malignancies. Citation Format: Qing Lan, Lauren M. Hurwitz, John K. Chan, Tai Hing Lam, Kexin Chen, Yok Lam Kwong, Xu Caigang, Brian CH Chiu, Raymond Liang, Ip Dennis, Wei Hu, Bryan Bassig, Mark Purdue, Jun Xu, Sarah Locke, Sophia S. Wang, James R. Cerhan, Sonja Berndt, Jonathan N. Hofmann, Jianxin Shi, Kai Yu, Shahinaz Gadalla, Lisa J. McReynolds, Rena Jones, Hongji Dai, Zhangyan Lyu, Lugui Qiu, Wei Liu, Huilai Zhang, Xianhuo Wang, Lindsay M. Morton, Stephen Chanock, Martha Linet, Melissa C. Friesen, Roel Vermeulen, Nathaniel Rothman. A multi-center hospital-based case-control study of lymphoid and myeloid neoplasms (AsiaLymph): Study design and initial findings for education, body mass index, and family history [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 5051.