Graham, Rondell P.

Abstract Despite advances in surgery and chemotherapy, patients with resectable pancreatic ductal adenocarcinoma (PDAC) face high recurrence rates (40-60%). Total neoadjuvant therapy (TNT) seeks to enhance margin-negative resections (R0) and address micrometastatic disease, with potential survival benefits. Yet, TNT still results in high recurrence rates, with 1-, 2-, and 3-year disease-free survival (DFS) at 65%, 48%, and 32%, respectively. Residual cancer extent is a key prognostic factor in PDAC after TNT. Tumor response scores show favorable outcomes only for patients achieving near-complete or complete pathologic response. In a cohort of 511 PDAC patients treated with TNT and surgery (2015-2022), improved DFS was only found in those with a complete pathologic response (pCR, 7.8%). This highlights the limited use of current response scoring in stratifying post-TNT patient risk, underscoring the need to better understand disease persistence and relapse. Here we developed a quantitative method to assess residual cancer. Using a digital pathology H&E library from 2180 PDAC cases (2000-2022), an AI platform segmented tissue into cancer epithelium, stroma, and normal regions. For 202 TNT-treated cases without pCR, we calculated a ratio of cancer area to stromal area from representative slides. Modeled after the histologic tumor regression grade (HTRG) system, cases were classified into <5%, 5-50%, or >50% cancer-to-stroma ratios. However, this measure alone showed no difference in DFS, suggesting tumor burden cannot fully explain differences in clinical outcomes. We hypothesized that the spatial organization of the tumor microenvironment (TME) is a determinant of patient outcomes. Drawing from landscape ecology, which studies spatial properties of ecosystems, we applied similar principles to analyze tissue architecture. Using an AI-based two-class segmentation of whole-slide images, we assessed spatial features like patch area, perimeter, density, and Shannon diversity. Python’s ‘pylandstats’ package calculated 21 spatial metrics, reduced via stepwise Cox regression using the Akaike information criterion (AIC). The final model included mean Shannon diversity, mean and standard deviation of stromal patch areas, and stromal patch density. Kaplan-Meier analysis of 202 samples stratified model predictions into quartiles, correlating with high (mean DFS 10.85 months), medium (19.32 months), and low risk (25.92 months) groups. These findings aligned with model expectations, underscoring the potential of spatial TME analysis to predict outcomes. Citation Format: Andrea V. Maraone, Zafar Siddiqui, Luis H. Cisneros, Merih D. Toruner, Mark Truty, Cornelius Thiels, Robert R. McWilliams, Steven R. Alberts, Rondell P. Graham, Carlo Maley, Qian Shi, Martin E. Fernandez-Zapico, Ryan M. Carr. Tumor spatial configuration predicts pancreatic cancer recurrence risk after total neoadjuvant therapy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 5070.

Abstract Background: Pancreatic cancer is a frequently fatal disease that is often diagnosed at a late stage associated with a poor prognosis and an increasing impact on cancer related mortality. Despite advances, chemotherapy remains the mainstem of therapeutic interventions for this disease and most of the currently identified molecular alterations are not currently targetable. Therefore, novel biomarkers and strategies are necessary to impact patient outcomes. Claudin-18.2 (CLDN18.2) is a splice variant of the Claudin 18 family and is a component of tight junctions, which seal the paracellular spaces of epithelial and endothelial cells. CLDN18.2 is frequently retained in malignant gastric tissue and overexpressed in pancreatic and gastroesophageal adenocarcinomas. CLDN18.2 overexpression at the message and protein levels has been observed in pre-clinical studies involving pancreatic cancer cell lines and in tumors from pancreatic cancer patients. The goal of this study was to obtain a deeper understanding of CLDN18.2 positivity and CLDN18.2 expression and the associated clinical prevalence, prognostic implications, patient clinical characteristics and relationship with other biomarkers in pancreatic ductal adenocarcinoma (PDAC) patient samples. Design: This was a single institution/reference laboratory based retrospective study including as many as 320 patients, in which CLDN18.2 positivity as defined by ≥ 75% of the tumor cells expressing moderate to strong membranous staining was examined. The goals were to evaluate CLDN18.2 in tumor samples by immunohistochemistry for prevalence in pancreatic adenocarcinoma patients. Claudin 18.2 expression was assessed by the Ventana (43-14A) Class I IVD that was purchased from Ventana/Roche Diagnostics. In addition to prevalence, CLDN18.2 was evaluated with respect to the clinicopathological features of patients, patient prognosis, adjusted for clinical covariables, response to treatments and relationship to patient demographics, tumor grade and disease status. Patient information has been obtained from electronic medical records and research database to establish clinical course (survival status, metastatic sites, etc.) along with patient demographics. Conclusions: CLDN18.2 is a novel potentially actionable biomarker in patients with advanced and metastatic pancreatic adenocarcinoma. This study provides important clinical information regarding CLDN18.2 expression in these patients that is important to consider when evaluating biomarkers in these patients. Funding for the study was provided by Astellas Pharma Global Development, Inc. Citation Format: Allison Kerper, Robert H. Getzenberg, Rondell P. Graham. Real world clinical characteristics of claudin 18.2 expression in patients with pancreatic adenocarcinoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 5915.

Spatial heterogeneity of liver viscoelasticity, measured at three-dimensional vector MR elastography, positively correlated with metabolic dysfunction–associated steatotic liver disease progression.

ABSTRACT Metabolic-associated steatohepatitis-driven hepatocellular carcinoma (MASH-HCC) incidence is rapidly rising worldwide. Lipid metabolic reprogramming is a hallmark of solid tumors to satisfy cancer high metabolic demand. However, it may confer sensitivity to ferroptosis, a cell death mode driven by iron-dependent lipid peroxidation. In this report, we describe the lipid metabolic landscape in MASH-HCC and characterize long chain acyl-CoA synthetases (ACSLs), a family of enzymes involved in synthesis of cellular lipids. Bulk RNA-sequencing, single-cell RNA-sequencing, spatial transcriptomics and immunohistochemistry analyses of human MASH-HCC were integrated to identify differentially expressed lipid metabolism genes. Ferroptosis in vitro was assessed in human HCC cell lines. A characterization of ACSLs was also conducted at the single-cell level in a diet-induced experimental murine model of MASH-HCC. Our analysis revealed that in human MASH-HCC, ACSLs exhibit a heterogeneous expression, with ACSL4 notably enriched in tumor tissues, contrasting with ACSL5 upregulation in non-cancerous MASH. We identified a unique lipid metabolic gene signature of MASH-HCC, which included genes associated with ferroptosis vulnerability. In vitro , ACSL4 upregulation was associated with increased ferroptosis sensitivity in human HCC cell lines. Lastly, single-cell RNA-sequencing revealed elevated ACSL4 expression in immune cells in a murine MASH-HCC model, suggesting a role of ACSL4 in shaping the tumor immune microenvironment. Overall, this report offers new insights into lipid metabolic landscape and ferroptosis sensitivity for novel MASH-HCC treatments. GRAPHICAL ABSTRACT

ABSTRACT EPHA2 is a receptor tyrosine kinase highly expressed in many cancers. By analyzing cancer patient databases for mutations in the EPHA2 coding sequence, we found that cholangiocarcinoma (a hepatobiliary cancer with dismal prognosis) exhibits a uniquely high incidence of EPHA2 mutations. To deUine the functional signiUicance of these mutations, we generated representative EPHA2 mutants and monitored major receptor autophosphorylation sites as indicative of kinase activity-dependent signal transduction (known as forward signaling). We found that missense mutations in the ligand-binding domain abrogate ligand binding and ligand-induced EPHA2 tyrosine phosphorylation, while most missense mutations in the kinase domain abrogate kinase activity. We detected less pronounced effects of missense mutations in other domains, which vary depending on the phosphosite, suggesting that these EPHA2 mutations might differentially affect (or bias) different downstream signaling pathways. Other EPHA2 mutations introduce early stop codons and encode receptor truncated forms lacking all or part of the kinase domain. We also found that an EPHA2 secreted truncated form, a transmembrane truncated form, and a full-length kinase inactive mutant can inhibit tyrosine phosphorylation of co-expressed EPHA2 wild-type. Taken together, these data suggest that mutations interfering with EPHA2 forward signaling facilitate cholangiocarcinoma development. We indeed obtained evidence that an EPHA2 kinase inactive mutant, but not EPHA2 wild-type, can induce proliferative masses consistent with well differentiated cholangiocarcinoma in a validated mouse model of cholangiocarcinogenesis. Taken together, our Uindings suggest that EPHA2 is a driver gene in cholangiocarcinoma and that its forward signaling has tumor suppressor activity.

Abstract Purpose: Digital proteomic profiling was performed to identify spatial context in relationship to patient response and survival after anti–PD-1 therapy in metastatic colorectal cancer. Experimental Design: Primary colorectal cancers with deficient mismatch repair from patients treated with anti–PD-1 antibodies were analyzed (N = 30) using digital spatial profiling (GeoMx nCounter). At the invasive margin, 71 proteins were profiled in 10 regions of interest/slide that were segmented into 3 compartments labeled with pan-cytokeratin (epithelia), CD45 (stromal cells), and SYTO13 (nuclei). In an independent cohort (n = 13), digital spatial profiling data and single-cell transcriptomic data were analyzed. Differential protein abundance, after Benjamini–Hochberg correction, was examined by clinical response and progression-free survival (PFS) using multivariable Cox regression. Results: Protein abundance varied significantly between epithelial and stromal compartments. Nonresponders to anti–PD-1 showed higher fibronectin and smooth muscle actin abundance in the epithelial compartment that was associated with significantly shorter PFS (adjusted HR: 6.49 and 4.52, respectively; P < 0.05). In CD45+ stroma, increased expression of proteins related to T cells (CD3 and CD4), NK cells (CD56), antigen presentation (CD40), immune activation (CD27, ICOS), and apoptosis (GZMA) were found in responders (vs nonresponders) to anti–PD-1; each marker was significantly associated with longer patient PFS (0.02 < adjusted HR < 0.17; P < 0.05). In a separate cohort, consistent results by compartment were found for fibronectin and CD56. Gene expression data revealed that fibronectin and smooth muscle actin were primarily derived from cancer-associated fibroblasts. Conclusions: Spatially resolved protein profiles within microenvironments of deficient mismatch repair colorectal cancers can influence patient response and survival after anti–PD-1, highlighting their potential clinical significance.

Abstract Given the heterogeneous clinical behavior of pancreatic neuroendocrine tumors (pNETs), improved prognostic markers are needed to guide management and post‐resection surveillance. Patients who underwent resection of large (≥3 cm) sporadic well‐differentiated pNETs from 2000 to 2019 were identified. The Ki‐67 proliferation index was determined using immunohistochemistry, and alternative lengthening of telomeres (ALT) status was assessed using fluorescence in situ hybridization. Recurrence‐free and overall survival were estimated using Kaplan–Meier analysis. Multivariable Cox regression analysis evaluated factors associated with recurrence‐free survival. A total of 106 patients were identified. ALT was positive in 57 (54%) and negative in 49 (46%). Ki‐67 was ≥3% in 74 (70%) and <3% in 32 (30%). Tumors with Ki‐67 ≥3% were more likely to be ALT positive (61% vs. 38%, p = .046). Stratifying by ALT status and Ki‐67 proliferation index, median recurrence‐free survival was 4.6 years for patients with ALT‐positive/Ki‐67 ≥3% tumors, 3.1 years for patients with ALT‐positive/Ki‐67 <3% tumors, 12.4 years for patients with ALT‐negative/Ki‐67 ≥3% tumors, and 20.2 years for patients with ALT‐negative/Ki‐67 <3% tumors ( p < .001). Initial recurrence was distant in 82% and locoregional in 18%. Across all groups, overall survival was similar ( p = .19). In multivariable analysis, advanced age, ALT positivity, perineural invasion, and lymph node metastases were associated with increased recurrence risk (all p < .05). ALT and Ki‐67 provide complementary information on post‐resection recurrence risk, which can guide subsequent surveillance and management strategies. These data support the incorporation of ALT testing into routine clinical practice.

AimsWe compiled two cohorts of colorectal adenosquamous carcinoma (ASC) to describe its histologic and molecular aspects using modern parameters to compare them with literature reports using meta‐analysis of cohorts and individual case series.Methods and ResultsWe identified 53 colorectal ASC from 19 North American academic medical centres, in addition to national database reports on 94 Dutch cases. We analysed available clinical, histologic, and immunohistochemical features and patient outcome. ASC comprised 0.02% of colorectal cancers in the Dutch database. The median cohort patient ages at resection were 65 and 69 years (North American and Dutch cohorts, respectively), with a roughly equal male:female ratio. The squamous component represented between 5% and 95% of each tumour, with a median of 50%. Tumour‐infiltrating lymphocytes (TILs) were generally low (66%), and tumour budding was often Bd1 (64%). Most cases were pT3 (55%) or pT4 (40%), with nodal metastases in more than half (58%). Twenty‐three cases (43%) metastasized distantly, most commonly to the liver. Mismatch repair (MMR) deficiency was identified in 34% of the cases. Follow‐up was available for 48 patients; 13 (27%) had recurrent disease and 29 (60%) died. A total of 31 patients progressed, with median time to progression of 18 months. Available data for the Dutch cohort revealed largely similar findings, as did review of cases in the literature.ConclusionColorectal ASC usually presents at an advanced stage. Despite high rates of MMR deficiency and low tumour budding, TILs were generally low, and there is a high recurrence rate and poor prognosis.

270 Background: While immune checkpoint blockade (ICB) is used to treat metastatic deficient mismatch repair (dMMR) colorectal cancers (CRCs), such tumors are heterogeneous and resistance to ICB is frequent. To date, biomarkers that can consistently predict patient response to ICB are lacking. We determined whether biomarkers that integrate tumor and immune-related molecular mechanisms could predict ICB response. Methods: Consecutive patients with metastatic dMMR CRCs (N=32) who had been treated with anti-PD-1 therapy were profiled using a validated immune-enhanced exome and transcriptome platform (ImmunoID NeXT, Personalis) using tumor and paired normal tissue. This platform provides antigen-specificity of T or B cells, indicated by the TCR or BCR respectively, neoantigen prediction, tumor mutation burden (TMB), immune profile, HLA allele-specific loss of heterogeneity (LOH) and a Neoantigen Presentation Score (NEOPS) that combines a neoantigen prediction tool with mechanisms of immune evasion. Tumor variables were analyzed in relationship to objective tumor response, and a Cox proportional hazards model was fit to predict progression-free survival (PFS) or overall survival (OS). Results: In CRCs, TCR and BCR repertoire diversity were each significantly increased in tumors from responders versus those with stable or progressive disease after anti-PD-1 treatment (p=0.015 and p=0.048, respectively). Neoantigen burden score, TMB, and HLA allele-specific LOH (reflects loss of antigen presentation machinery) were each unable to distinguish responders from nonresponders. An immune checkpoint gene expression signature was increased in nonresponders relative to responders (P= 0.005). Neoantigen presentation score (NEOPS) was prognostic when adjusted for tumor grade only for OS (p=0.048). Conclusions: The T-cell and B-cell receptor (TCR/BCR) repertoire, as well as NEOPS, were each significantly predictive of favorable response to anti-PD-1 therapy in dMMR tumors. In addition, an immune checkpoint gene expression signature was predictive of resistance to ICB.

251 Background: The development of colorectal liver metastases (CRLM) is associated with poor prognosis, and recent data suggest that metastasis to the liver is associated with resistance to immunotherapy. We characterized the microenvironment of primary colorectal carcinomas (CRCs) relative to their synchronous CRLM using a validated segmentation algorithm that quantifies 15 distinct morphologic tumor features. Methods: Adult CRC patients with synchronous CRLM (N=57) at Mayo Clinic were identified from the electronic health record using Epic Slicer Dicer and from an internal database. Tumor H&E sections were digitized and reviewed for quality control (RKP). QuantCRC (Aiforia, Inc) was applied to digitized images to extract 15 GI pathologist pre-defined morphological features. Tumor features were compared between primaries and CRLM using the Kruskal–Wallis test. The project was approved by the Mayo Clinic Institutional Review Board. Results: The study included 57 patients (median age 59 years [IQR 50, 73], 51% female) with CRC primaries and synchronous CRLM. Among primaries, 20 (35%) were right-sided and 37 (65%) were left-sided. QuantCRC identified 6 of 15 morphological features that differed significantly between primaries and their CRLM, including reduced stroma, more high-grade and necrosis, and a higher tumor: stromal ratio (TSR) in CRLM (Table). The increase in TIL density in CRLM vs the primary tumor was of borderline significance (p =0.053). Among patients with left-sided primary tumors, their CRLM had significantly higher TSR, percent high-grade, percent necrosis, and TIL density compared to the primary (all p values ≤ 0.02). Among right sided primaries, CRLM had a significantly reduced percent mature stroma (p=0.034) whereas percent necrosis was increased (p =0.01). Conclusions: Using deep learning, we identified tumor morphological features that differed significantly between primary CRC and their synchronous CRLM. This included higher TSR in CRLMs compared to primaries that is associated with epithelial-mesenchymal transition and has been shown to contribute to treatment resistance. Analysis of tumor morphological features with patient prognosis is ongoing. Deep learning-derived morphological features of primary CRC and CRLM. Morphological feature (Median) Primary tumor Liver metastasis P-value Tumor-Stroma Ratio 0.78 1.50 0.008 % High-grade 23.08 31.72 0.005 TIL Count 44.36 60.20 0.053 % Necrosis 6.35 19.32 <0.001 %Signet Ring Cells 0.25 0.13 0.088 TB/PDC 0.89 1.13 0.484 % Stroma 54.89 40.98 <0.001 % Immature Stroma 46.76 34.13 <0.001 % Mature Stroma 6.27 5.08 0.030