Background: Prostate cancer is a heterogeneous disease with variable clinical outcomes. If localized, the patient may be cured. However, prostate cancer is lethal if recurrence/progression to metastatic castrate resistant disease occurs. Thus, there is an unmet need to further understand the molecular underpinnings of this progression. Epidemiologic studies show that increased risk of developing and dying from prostate cancer has been associated with elevated serum IGF-1 levels, hyperinsulinemia and metabolic syndrome. Alterations in insulin pathway genes, such as PTEN, FOXO, and PIK3CA, are mutated in up to 32%, 15%, and 11% of localized prostate tumors, respectively. We aimed to further characterize expression of insulin pathway genes in localized prostate cancers in an effort to (1) provide insights into potential mechanisms of progression to metastatic disease and (2) try to further enrich for those prostate tumors that portend worse survival outcomes. Methods: Using the multi-institutional Oncology Research Information Exchange Network (ORIEN) database, gene expression data was analyzed from localized prostate cancer tumors. The raw counts were first normalized, and 176 genes related to the insulin receptor and its downstream pathways were then subset and used for clustering using the non-negative matrix factorization (NMF). The NMF cluster analysis was performed in an attempt to separate gene expression into two groups. Gene Set Enrichment Analysis (GSEA) was then performed between the two groups that had been separated by cluster analysis to determine homology between other GSEA sets. Kaplan-Meier curves were used to assess median overall survival. Cox analysis was performed to generate the adjusted KM curve. Mediation analysis was conducted to determine the relationship between cluster status, TN stage, and survival. Results: Cluster analysis revealed two distinct groups of insulin gene expression, cluster 1 (n = 96) and cluster 2 (n = 337). Compared with cluster 2, cluster 1 consisted of decreased expression of PTEN (p < 0.001) and PIK3R1 (p < 0.001), along with increases in the expression of AKT1 (p < 0.001), IRS1/2 (p < 0.001), FASN (p < 0.001), IGFBP2 (p < 0.001), and MTOR (p < 0.001). GSEA analysis revealed changes in lipid metabolism and WNT secretion pathways in cluster 1. Cluster 2 GSEA showed pathway changes related to DNA damage repair and testosterone. Patient characteristics between clusters differed significantly in the T and N stages of tumor but not in other ways. In unadjusted analysis, median overall survival was estimated at 117 months and 232 months for cluster 1 and cluster 2, respectively (p < 0.05). The proportion of patients who went on to develop metastases (p < 0.05) or need chemotherapy (p < 0.05) was increased in cluster 1 compared to cluster 2. Repeat survival analysis adjusted for confounders (T stage, N stage, age at diagnosis, pathologic grade) showed no difference in survival between clusters. Mediation analysis showed that the contribution of cluster status to survival was independent of T or N stage. Conclusions: A subset of localized prostate cancer patients demonstrated linked insulin pathway changes that are consistent with prior studies describing a pattern of insulin dysregulation. Though the group characterized by insulin dysregulation initially showed worse survival outcomes, this difference disappeared when controlling for confounders. Though baseline differences in tumor stage seemed to most readily explain the difference in survival between clusters, mediation analysis showed that the effect of cluster status on survival was independent of tumor stage. This suggests that other confounders, such as pathologic grade or baseline age, may explain the survival difference.

e12760 Background: Advances in molecular classification have led to subtype-specific treatments for breast cancer, improving outcomes while revealing distinct recurrence patterns. This study compares outcomes and failure patterns between triple-negative breast cancer (TNBC) and estrogen receptor–positive (ER-positive) disease. Methods: We conducted a retrospective analysis using whole dataset of the Oncology Research Information Exchange Network (ORIEN), including women with non-metastatic breast cancer. Patients were categorized as TNBC or ER-positive. Outcomes included overall survival (OS), disease-free survival (DFS), locoregional control, and patterns of failure. Results: A total of 1,277 patients with stage I–III breast cancer were included, comprising 276 patients with TNBC and 1,001 with ER-positive disease. The median age was 52 years (interquartile range [IQR], 44–62). Most patients were White (83.3%) and non-Hispanic (84.5%). HER2 was positive in 13.4% of ER-positive disease. Compared with TNBC, ER-positive patients more frequently presented with clinical stage I (37.8% vs. 31.9%) and stage II disease (43.0% vs. 39.9%). In contrast, clinical stage III disease was more common among patients with TNBC than those with ER-positive disease (15.2% vs. 8.0%). Neoadjuvant systemic therapy was administered to 154 patients, including 62 (22.5%) with TNBC and 92 (9.2%) with ER-positive disease. Pathological complete response after neoadjuvant systemic therapy was observed in 14 (9.1%) patients (6 TNBC and 8 ER-positive). Adjuvant systemic therapy was delivered to 641 patients, including 108 (39.1%) with TNBC and 533 (53.2%) with ER-positive disease. Among the ER-positive patients who received adjuvant systemic therapy (n = 533), 498 (93.4%) received adjuvant endocrine therapy. Radiation therapy was delivered to approximately two-thirds of patients in both groups. Among those treated with radiation, 83.7% received whole-breast or chest wall irradiation with nodal coverage. Lymph node irradiation was administered in 36.8% of patients, not administered in 33.7%, and was unknown in 29.5%. With a median follow-up of 75.2 months (range, 0–393.7 months), TNBC was associated with significantly worse overall survival (hazard ratio [HR] 2.85, 95% confidence interval [CI] 2.04–3.96), disease-free survival (HR 2.10, 95% CI 1.56–2.83), and distant metastasis–free survival (HR 2.28, 95% CI 1.25–4.16) compared with ER-positive disease. No significant differences were observed in local control (HR 1.17, 95% CI 0.51–2.70) or regional control (HR 2.28, 95% CI 0.85–6.11) between the two groups. Conclusions: TNBC was associated with significantly worse survival outcomes and higher rates of distant failure compared with ER-positive disease. These results highlight distant metastasis as the dominant pattern of failure in TNBC and support the need for more effective systemic therapies.

752 Background: Pancreatic ductal adenocarcinoma (PDAC) is a deadly malignancy with poor prognosis. A transcriptomic based algorithm called Purity Independent Subtyping of Tumors (PurIST) relies on expression profiling of a defined set of genes to stratify PDAC into classical and basal-like molecular subtypes. Though clinical trials are evaluating PurIST as a tool to guide chemotherapy clinical decisions, no study has investigated the utility of PurIST in the Appalachian population, which has many known health disparities. In this study, we apply PurIST to Appalachian and non-Appalachian patient cohorts and assess demographic variables between groups that may influence treatment outcomes. Methods: This 13-site study includes 780 PDAC patients via the Total Cancer Care protocol and was conducted via the Oncology Research Information Exchange Network (ORIEN) with Aster Insights. Patients were divided into two cohorts following the guidelines of the Appalachian Regional Commission and analyzed for differences in PurIST classification, tobacco use, alcohol use, and age at last contact. Results: Of the 780 PDAC patients assessed using PurIST, 562 (72.05%) were classical (497 strong classical; 165 Appalachian; 397 non-Appalachian), 63 (8.07%) were basal (22 strong basal; 13 Appalachian; 50 non-Appalachian), and 155 (19.87%) did not fall into a dichotomous subtype (59 Appalachian; 97 non-Appalachian). Significantly fewer individuals were able to be classified with PurIST in Appalachia (P=0.02321) and the Appalachian cohort trended toward increased classical subtype (P=0.1457). Appalachian patients were more likely to have a history of smoking (59.6% vs 49.9%; P=0.01756) and less likely to report alcohol consumption (50.5% vs 65.6%; P=0.000229). A trend toward more basal-like tumors was identified in advanced age individuals (P=0.02542; 20.5% in ages 85+ compared to 9.4% in individuals younger than 85), but it was not significant after statistical correction. Conclusions: We conducted the largest comparison of the PurIST algorithm in Appalachia to date and found increased expression of the classical subtype across all populations, with a trend toward higher classical and unclassifiable tumors within Appalachia. If PurIST is to be considered either as a prognostic or decision-making tool for PDAC patients, we must account for potential variability in different populations.

4194 Background: Mucin 5AC (MUC5AC) is aberrantly expressed in pancreatic ductal adenocarcinoma (PDAC) and associated with tumor progression; however, its role in altering the tumor immune microenvironment (TIME) and immunotherapy-relevant pathways is poorly understood. Thus, we evaluated whether MUC5AC expression delineates PDAC into distinct transcriptional and immunologic subtypes based on TIME. Methods: Bulk RNA-seq data from 730 PDAC tumors (Oncology Research Information Exchange Network, ORIEN) and 360 normal pancreatic tissues (GTEx) were analyzed. Differential gene expression and pathway enrichment were assessed using Gene Ontology and KEGG analyses. Immune infiltration and stromal states were inferred using multiple complementary deconvolution algorithms (CIBERSORT and xCell). Tumors in the lowest quartile of MUC5AC expression (≤25th percentile; MUC5AC-low, MUC5AC-L) were compared with the remaining samples (MUC5AC-high, MUC5AC-H). Associations between MUC5AC expression and immune cell populations and immunomodulatory genes were evaluated by using correlations and group-based analyses with false discovery rate (FDR) corrections. Results: Principal component analysis demonstrated clear separation between PDAC tumors and normal pancreas and revealed distinct transcriptional programs between MUC5AC-H (N = 544) and MUC5AC-L (N = 186) tumors. MUC5AC-H tumors exhibited a profoundly immunosuppressive and stromal-remodeled TIME, including enrichment of M2 macrophages, regulatory T cells, activated natural killer ells, fibroblast-associated signatures, and significantly higher immune and stromal scores (multiple FDR ≤10⁻⁶ to ≤10⁻²⁸). In contrast, MUC5AC-L tumors showed relatively enrichment of immune-activating populations, including naïve B cells, dendritic cell subsets, resting CD4⁺ memory T cells, and CD8⁺ T cells (multiple FDR < 10⁻¹⁵). MUC5AC expression strongly correlated with immune checkpoint genes CD274 (PD-L1) and TIGIT , and with multiple suppressive immune populations (p < 0.01 to < 10⁻¹⁰). Conclusions: MUC5AC defines a transcriptionally distinct, immunosuppressive PDAC subtype characterized by stromal activation, macrophage/Treg enrichment, and immune checkpoint engagement. MUC5AC may serve as a biomarker for immune stratification and supports rational combination strategies integrating stromal targeting with immune checkpoint modulation in PDAC.

e16313 Background: Well-differentiated GEP-NETs comprise a heterogenous group of malignancies with variable behavior. While existing grading structures attempt to predict these behaviors, significant variability remains. The ORIEN database contains complementary clinical, genomic, and transcriptomic profiling, providing a novel source for correlating molecular data with clinical characteristics. Methods: mRNA expression profiles were assessed for correlation the presence (M1) or absence (M0) of metastatic disease at diagnosis. The workflow consisted of: (1) Data preparation: RSEM gene expression data were processed, quality-controlled, and mapped to clinical metadata. (2) Differential gene expression: DESeq2 was used to identify significantly differentially expressed genes (FDR < 0.05). (3) Pathway enrichment analysis: Gene Ontology (GO), KEGG, and Reactome enrichment analyses were performed on all differentially expressed genes to identify biological pathways associated with metastatic status. Results: Across all GEP-NETs, 112 genes had significantly different expression levels for M1 (n = 5) vs. M0 (n = 24) stage, including 35 genes expressed at higher levels in M1 disease. In analysis limited to pNETs, 708 genes showed significantly different expression levels between M1 (n = 3) and M0 (n = 20) stages, including 155 genes expressed at higher levels in in M1 disease. In both the entire GEP-NET cohort and pNET subgroup analysis, the most frequently overexpressed Reactome pathways associated with M1 stage were related to the regulation of metaphase transition and DNA-protein (histone) interactions. Cancer testis antigen (CTA) genes were also frequently upregulated (Table 1). Conclusions: Differential mRNA expression data provides a novel means for better characterizing the molecular underpinnings of GEP-NETs. Specifically, cell cycle, histone-related, and CTA gene expression profiles may provide actionable findings on which to base downstream drug development for the prevention and/or treatment of metastatic disease. Most upregulated mRNA profiles by magnitude in metastatic GEP-NETs. Gene Log 2 Fold ∆ a Description / Function GAGE2A 7.41 CTA, chromatin regulator, anti-apoptosis RNF17 5.89 CTA, DNA damage response GTSF1 5.76 CTA, DNA methylation, retrotransposon regulation NBEAP1 5.19 Also called BCL8, little known normal physiologic function ORM1 4.53 Acute phase reactant, angiogenesis, immune evasion HIST2H3C 4.18 Histone H3 regulation, chromatin regulator LYZ 4.12 Lysozyme production, TME inflammation, immune evasion PBK 4.02 CTA, cell cycle regulation, histone phosphorylation, anti-apoptosis HIST2H3A 3.71 Histone H3 regulation, chromatin regulator a All p-values <0.0001.

Purpose:In a previous study, we trained, validated and tested models of endometrial cancer (EC) recurrence integrating clinical, genomic and pathological data from the Oncology Research Information Exchange Network (ORIEN). Preliminary studies also have demonstrated that bacterial communities may influence the risk of EC recurrence by altering the local environment within the upper female genital tract. The objective of this study was to evaluate whether extrinsic and environmental factors, including tumor-associated bacterial communities, tumor immune contexture and air pollution alongside clinical, pathologic and genomic features are associated with EC recurrence across clinically relevant risk groups. Patients and Methods:We performed a retrospective, multi-institution, case-control study with data from the ORIEN network EC dataset. Data was stratified into low-risk, FIGO grade 1 and 2, stage I (N = 329), high-risk, or FIGO grade 3 or stages II-IV (N = 324), and non-endometrioid histology (N = 239) groups. RNA and DNA were extracted from tumor specimens and processed to obtain the necessary genomic/metagenomic data. Genus level microbiome data were extracted and curated) from RNA sequencing using Kraken2, Bracken and exotic software packages. Risk of EC recurrence was evaluated by integrating microbiome and environmental data alongside existing clinical, pathological and genomic data using topic modelling with latent dirichlet allocation (LDA). Prediction models of EC recurrence were created using machine and deep learning analytics (ML and DL) with MATLAB apps and TensorFlow. Finally, performance of both topic and prediction models were externally validated in an independent EC dataset from TCGA. Results:The resulting models, analyzed with topic modelling, demonstrated the complexity of factors involved in recurrence of disease for EC. The components of the resulting topic models, and specifically the microbiome, changed when environmental factors, like air pollutants, were introduced in the model. In the low-risk EC group, microbes that were quite abundant in models before introducing environmental factors, were scarcely seen afterwards, like genera Thermothielavioides, Theileria, Rhizoctonia. Bacillus was the genus with higher per-topic probability within all risk groups, especially for low-risk EC (28%). Ozone (O3) was a resulting component of all risk groups' models. BMI was the sole informative clinical variable after data integration, and only present in the low-risk group. Resulting models from the high-risk and non-endometrioid groups included differential gene expressions: MMP13, S100A7, SMOC1, ACACA and ADD2, DLX5, SLCO2B1, NWD1 respectively. CNVs also were present in both low-risk and non-endometrioid groups, but their per-topic probabilities were low. The same was true for the immune contexture data. The components of the resulting topic models were used to train, validate and test prediction models of EC recurrence by risk groups. Performances of these models were excellent (@ 0.9). Despite some missing microbiome data in TCGA from resulting topic models, prediction models trained in the ORIEN set, had similar performances in TCGA testing set, with overlapping AUC 95% CIs. Conclusion:Both extrinsic factors (tumor-associated bacterial communities, tumor immune contexture and air pollution) and intrinsic factors predict EC recurrence. The complexity of tumor and host factors influencing cancer relapses underscore the need for more individualized prediction models of disease outcomes.

Thyroid diseases are common and highly heritable. We performed a meta-analysis of genome-wide association studies from 19 biobanks for five thyroid diseases: thyroid cancer (ThC), benign nodular goiter, Graves’ disease, lymphocytic thyroiditis and primary hypothyroidism. We analyzed genetic association data from ~2.9 million genomes and identified 313 known and 570 new independent loci linked to thyroid diseases. We discovered genetic correlations between ThC, benign nodular goiter and autoimmune thyroid diseases ( rg = 0.16–0.97). Telomere maintenance genes contributed to benign and malignant thyroid nodular disease risk, whereas cell cycle, DNA repair and damage response genes were associated with ThC. We propose a paradigm that explains genetic predisposition to benign and malignant thyroid nodules. We found polygenic risk score associations with ThC risk of structural disease recurrence, tumor size, multifocality, lymph node metastases and extranodal extension. Polygenic risk scores identified individuals with aggressive ThC in a biobank, creating an opportunity for genetically informed population screening.

e16326 Background: Well- to moderately differentiated gastrointestinal and pancreatic neuroendocrine tumors (NETs) show substantial heterogeneity in clinical outcomes. While tumor-related characteristics are established prognostic factors, the impact of patient-level comorbidities, metabolic factors, and sociodemographic characteristics on overall survival (OS) remains incompletely described. We evaluated OS across clinical, metabolic, renal, and demographic subgroups in a real- world cohort of patients with well- to moderately differentiated NETs. Methods: We conducted a retrospective cohort study using a harmonized clinical dataset spanning diagnosis, physical assessment, laboratory, and demographic domains. The cohort included adult patients with well- to moderately differentiated NETs of pancreatic and gastrointestinal origin; poorly differentiated tumors were excluded. Baseline characteristics, including body mass index (BMI, categorized as < 25 vs. ≥25 kg/m²), smoking status (ever vs. never), age at primary diagnosis, and race/ethnicity, were extracted. Advanced kidney disease (AKD) was defined as serum creatinine > 1.5 mg/dL on ≥2 occasions. OS was estimated with the Kaplan-Meier method, and differences were assessed with the log-rank test. Results: The cohort included 132 patients (Oncology Research Information Exchange Network, ORIEN). Overall survival was significantly worse in patients aged ≥50 years, with a shorter median OS than in patients younger than 50 years. For patients who were 50 or older at primary diagnosis, the median overall survival was 14.46 years, whereas for those younger than 50, the median OS was not reached(p = 0.046). There was no significant difference in median OS based on other clinical factors, including BMI, smoking status, advanced kidney disease, and race/ethnicity. Conclusions: In this real-world analysis of patients with GEP-NETs, age was the dominant predictor of overall survival.Traditional metabolic and renal risk factors did not demonstrate significant prognostic value in this cohort, suggesting that tumor biology or age-related background mortality may outweigh these comorbidities in this setting. Future studies utilizing larger, multi-institutional datasets are warranted to perform multivariate adjustments and isolate disease-specific survival outcomes.

Over a decade ago, collaborating academic cancer centers formed the Oncology Research Information Exchange Network® (ORIEN) to develop a patient-driven, federated infrastructure for oncology research. Aster Insights is the network’s operational, commercial, and research partner. Together ORIEN and Aster Insights have built a unique multimodal dataset on the active engagement and consent of patients who opted into the Total Cancer Care® (TCC) protocol to contribute their data and biospecimens for research. Over 400,000 cancer patients have been enrolled in TCC, >32,500 of which have an in silico “Avatar” generated to represent their individual patient experience and molecular profile to support a broad range of network and industry research use cases. We provide an overview of ORIEN’s evolution, demonstrate the power of our data resources through a landmark analysis of >37,000 tumors across all cancer types collected for the Avatar program, and provide a vision for ORIEN to fuel collaborative research.

Longitudinal data analysis of the patient’s treatment course is critical to uncovering variables that influence outcomes. However, existing tools have significant limitations in integrating multilayered time-series data, particularly in linking treatment events with survival outcomes. Here, we developed ShinyEvents, a web-based framework for complex longitudinal data analysis. ShinyEvents allows users to upload data and generate interactive timelines of clinical events, enabling cohort-level analyses such as treatment clustering and endpoint assignment. It also provides informative cohort visualizations, such as a Sankey diagram of the treatment line and a Swimmer diagram of the clinical course. Finally, our tool can infer real-world progression-free survival (rwPFS) based on user-defined endpoints and perform Kaplan-Meier and Cox proportional hazards regression analysis. With these features, the tool can then associate treatment lines with clinical outcomes. As a case study, we analyzed Moffitt patients with muscle-invasive bladder cancer treated with neoadjuvant chemotherapy followed by surgery. Patients treated with cisplatin and gemcitabine exhibited more favorable rwPFS and overall survival, which is consistent with prior reports. Altogether, ShinyEvents provides a unified framework for integrating longitudinal real-world data with survival analytics, fostering transparent and reproducible collaboration between clinicians and data scientists. A live demo is available at https://shawlab-moffitt.shinyapps.io/shinyevents/.

Purpose:SWI/SNF (BAF) chromatin remodeling complex alterations are common in urothelial carcinoma, yet no biomarker-directed therapeutic strategies have been established for this population. We investigated whether BAF alterations delineate a biologically distinct, therapeutically actionable urothelial carcinoma subtype. Experimental Design:We performed integrative genomic and transcriptomic analyses of 792 urothelial carcinoma tumors from the Oncology Research Information Exchange Network (ORIEN) and validated findings in the TCGA-BLCA cohort. Mechanistic studies incorporated RNA sequencing and ATAC-seq following histone deacetylase (HDAC) inhibition. Functional dependencies were assessed using patient-derived xenograft organoids and cell line models. Clinical relevance was explored in a biomarker-enriched investigator-initiated trial. Results:Approximately half of urothelial carcinoma tumors exhibited BAF alterations, defining a previously unrecognized chromatin-altered molecular subtype characterized by activation of proliferative programs, loss of lineage identity, and altered metabolic signaling. This subtype was enriched for transcriptomic programs associated with HDAC inhibitor sensitivity and depleted of HDAC inhibitor resistance signatures. Mechanistically, HDAC inhibition induced widespread chromatin remodeling with reduced accessibility at AP-1 and TEAD-associated regions, and downregulation of E2F- and MYC-driven transcriptional networks. Functional studies confirmed enhanced HDAC inhibition sensitivity in ARID1A -mutated cell lines and a patient-derived organoid model. Early clinical observations demonstrated a durable responder treated with HDAC inhibitors and immunotherapy. Conclusions:BAF alterations define a chromatin-dependent tumor state in urothelial carcinoma that is selectively vulnerable to HDAC inhibition. Integrating genomic, epigenomic, functional, and early clinical evidence, these findings provide a rationale for biomarker-enriched clinical trials and HDAC inhibitor-based combination strategies in urothelial carcinoma.

Selected log2 fold change results for microbes found to be significantly enriched in hypoxic tumors in both mice and human subjects.

Modeling results for gene expression differences between responders and non-responders

Relationship between microbes and immune cell composition. A, All microbe-genus level abundances (columns) correlated to the deconvolved immune cell abundances (rows), with the strength of the association and its direction indicated by color in the TCGA (top row) and ORIEN (bottom row) datasets. Spearman correlations between the deconvolved abundance of (B) activated mast cells or (C) neutrophils and microbes in both the TCGA and ORIEN datasets. Concordant associations (significant in both and in the same direction) are shown in green, with select taxa labeled. There were no microbes significantly anticorrelated with activated mast cells or neutrophils.

577 Background: Approximately 30% of patients with ccRCC present with metastatic disease (stage IV) and close to half of patients with stage III disease will recur during follow-up surveillance. Two of the most common and morbid sites of metastatic development are brain and bone metastasis. Existing treatment guidelines do not recommend routine brain or bone directed imaging during either initial staging or surveillance in the absence of clinical signs or symptoms. A group of recurrently altered aberrant splice variants in primary ccRCC tumors associated with metastatic progression have previously been identified. Our study aims to investigate metastatic organotropism of the FGD1 -splice variant ( FGD1 -SV) to refine risk stratification and therapeutic decision-making as cabozantinib is drug with activity at these disease sites. Methods: This study leveraged the ORIEN AVATAR network, a data-sharing alliance of 18 NCI-designated cancer centers, to evaluate the presence of FGD1 -SV in a cohort of 1,001 clear cell renal cell carcinoma (ccRCC) patients using bulk RNA-sequencing (RNA-seq). Samples with three or more FGD1 -SV reads were classified as positive. Clinical and survival data were collected, and Kaplan-Meier curves were generated. Odds ratios (ORs) evaluated the presence of FGD1 -SV on brain and bone metastases, while hazard ratios (HRs) assessed association with cabozantinib response. Of 1,037 RNA-seq samples, 84 were metastatic tumor specimens. The relationship between FGD1 -SV positivity and metastasis to common sites was assessed with Fisher's exact test. Results: Brain and bone metastases demonstrated the highest FGD1 -SV positivity (50% and 44%, respectively). A grouped comparison of brain and bone metastases versus all other common metastatic sites (adrenal, lymph node, pancreas, lung, and liver) revealed significant enrichment of FGD1 -SV in these metastases (OR 5.33, 95% CI [1.64 - 18.42], p=0.002). FGD1 -SV positivity in primary tumors was associated with greater risk of recurrence after surgical resection (p=0.0029). Furthermore, FGD1 -SV positive tumors were associated with poor survival when not treated with cabozantinib (HR 2.04, 95% CI [1.20 - 3.45], p=0.01). Conclusions: FGD1 -SV positivity is significantly associated with brain and bone metastatic organotropism in ccRCC. Our findings warrant prospective studies to validate these results and investigate the integration of FGD1 -SV into clinical decision-making, potentially guiding surveillance and therapeutic approaches in high-risk and metastatic patients.

BACKGROUND:Mutational landscape is prognostic in colorectal cancer (CRC). Rat sarcoma (RAS) oncogenes, such as KRAS and NRAS, with driver mutations, portend poor survival outcomes, whereas pathologic mutations in HRAS are extremely rare, and their prognostic value remains uncertain. METHODS:This retrospective study analyzed the Oncology Research Information Exchange Network (ORIEN) alliance tumor RNA-Seq data in Stages II and III CRC to investigate the association between RAS gene expression and survival outcomes. RESULTS:High transcript levels of HRAS were associated with superior overall survival (OS). The high HRAS-associated OS benefit was most pronounced in patients with right-sided primary expressing low KRAS transcript levels in the absence of pathologic KRAS mutations. CONCLUSIONS:Contrary to the notion that RAS family genes are proto-oncogenic, this study demonstrates that high HRAS transcript levels are associated with superior OS in Stages II and III CRC. The potential of HRAS as a prognostic biomarker should be explored further.

222 Background: Androgen indifferent prostate cancer (AIPC) is increasingly common and particularly lethal. Data describing these tumors are sparse and AIPC remains a poorly understood malignancy. This study aims to characterize the clinical and genomic features of AIPC. Our work ultimately seeks to identify biomarkers with diagnostic and therapeutic potential. Methods: Utilizing the Oncology Research Information Exchange Network (ORIEN) database, we queried all prostate cancer (PC) patients, identified metastatic castrate resistant prostate cancer (MCRPC) samples, and aimed to enrich for tumors with features of AIPC using previously described characteristics. Our AIPC cohort included three subgroups: aggressive variant prostate cancer (AVPC) defined as having alterations in at least two of TP53, RB1, PTEN; neuroendocrine PC (NEPC) defined as small cell histology or NEPC signature score ≥ 0.25 (1); and double-negative PC (DNPC), defined as non-NEPC patients with low AR expression/AR signaling score. We compared clinical characteristics and genomic analysis of AIPC vs non-AIPC samples in patients who developed MCRPC. Clinical analysis was done using Wilcoxon rank sum test or Fisher's exact test. Gene expression analysis was performed using DESeq2 and GSEA. Results: Of 1,496 total PC patients available for analysis, we identified 323 (22%) as MCRPC. Of those, 39 (12%) met AIPC criteria (17 AVPC, 13 NEPC, 9 DNPC) and 284 (88%) were non-AIPC. Median age at diagnosis for AIPC was 62 years and 85% were white, compared to 62 years and 87% for non-AIPC. Fifty-seven percent of AIPC patients had ECOG ≥1 at diagnosis vs 16% of non-AIPC. Forty-three percent of AIPC patients had de novo metastatic disease vs 15% for non-AIPC (p=0.003). TMPRSS2-ERG gene fusions were found in a significantly higher proportion of AIPC samples vs non-AIPC (38.5% vs 16%, p=0.014). Homologous recombination deficiency (HRD) and tumor mutational burden (TMB) did not differ between cohorts, but microsatellite instability scores (MSI) were significantly higher in AIPC (p=0.019). Using Gene Set Enrichment Analysis (GSEA), we found that genes defining response to androgens and genes involved in oxidative phosphorylation were the most downregulated, whereas genes involved in epithelial mesenchymal transition (EMT), interferon response, and angiogenesis were significantly upregulated in AIPC vs non-AIPC samples. Conclusions: There was a significantly higher rate of de novo metastasis in the AIPC cohort. The downregulated androgen response and upregulated EMT pathways in AIPC suggest enrichment for androgen indifference with our methodology. Upregulated immune signaling and angiogenesis as well as higher MSI suggest opportunities for therapeutic investigation. Future directions include more focused in vitro and in vivo analysis to identify actionable targets. 1. Beltran H, et al. Nat Med . 2016;22(3):298-305. doi:10.1038/nm.4045.