Citations

Abstract Uveal melanoma (UM), a rare yet aggressive intraocular cancer with an estimated 5 cases per million people annually, presents a high metastatic risk and poor prognosis, highlighting an immeasurable need for novel diagnostic and therapeutic strategies. Not only is UM characterized by complex genetic and epigenetic aberrations, but it is also prone to deregulation of non-coding RNAs that could be used as therapeutic targets of diagnostic markers. In this study, we analyzed whole exome sequencing data from tumor and control samples of patients diagnosed with both uveal and cutaneous melanoma, focusing on a patient with a germline mutation in MBD4. Our framework, made up to work with long non-coding RNAs (lncRNAs), aims to flesh out these molecules' novel roles in disease etiology and progression. We identified several lncRNAs significantly associated with UM progression and computationally validated them in conjunction with microenvironmental factors in existing literature. LncRNAs are instructive in maintaining cellular homeostasis and, when deregulated, contribute to oncogenesis. Our results implied that the interaction between lncRNAs and other non-coding RNAs might be one of the most relevant mechanisms influencing UM pathogenesis. More interestingly, microenvironmental factors and lncRNA interplay are crucial in disease progression. The clinical relevance of our findings was tested using radiation treatment to evaluate potential therapeutic implications. Over and above that, we assessed the potential of lncRNAs as biomarkers and therapeutic targets in personalized medicine for UM patients to address ineffective treatments. The new lncRNA panels could serve as a new avenue for targeted therapy. With an in-depth understanding of the descent of molecular alterations and how lncRNAs interact within the tumor microenvironment, we can better size up potential strategies to cast off the challenges UM poses. Our work provides instructive insights into lncRNAs underpinnings of UM, emphasizing the significance of lncRNAs in tumor biology and their interactions within the microenvironment. We believe our findings will lay the groundwork for future therapeutic strategies and underscore the necessity for continued research to unravel the complex mechanisms in UM pathogenesis. Citation Format: Vasileios Alevizos, Nikitas Gerolimos, Sabrina Edralin, Clark Xu, Akebu Simasiku, Georgios Priniotakis, George A. Papakostas, Zongliang Yue. Unraveling long non-coding RNA dynamics in uveal melanoma: Insights into pathogenesis, biomarkers, and targeted therapies [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 2800.

Abstract Breast cancer risk assessment tools are widely used in clinical practice to guide decisions regarding screening timing and modality, lifestyle interventions, genetic testing, preventive therapy, and risk-reducing surgery. Although a number of tools designed for the general population are used in practice, they face a series of challenges including: (i) modest discriminatory ability due to lack of a unified model that flexibly incorporates a comprehensive set of risk factors; (ii) inability to produce subtype-specific risk estimates, relevant for tailoring screening and/or preventive strategies (e.g. prophylactic endocrine therapy that is effective only for hormone receptor positive tumors); (iii) lack of data from diverse populations to build models that work well for all women; and, (iv) scarcity of opportunities for comparative validation of different models. The Breast Cancer Risk Prediction Project (BCRPP), a collaboration currently including 20 prospective cohorts from the National Cancer Institute’s Cohort Consortium, is addressing these challenges by harmonizing, sharing, and analyzing data on over 1.8 million women, over 115, 000 of whom have developed breast cancer. The BCRPP is developing a comprehensive tool that will predict breast cancer risk, overall and by subtypes, across major racial and ethnic groups in the United States. This tool incorporates known and emerging risk factors (e.g. reproductive history, anthropometry, smoking, alcohol intake, medication use, mammography, family history, and measured genotypes). In parallel to model building and validation efforts, BCRPP investigators are developing privacy-preserving web and mobile interfaces to deliver the risk tools to patients and clinicians. Here we present descriptive data on participating cohorts (e.g. dates of enrollment, ages at enrollment, length of follow up, numbers of cases by subtype), summarize missing data patterns and our approach to model development and validation that accounts for heterogeneity in missingness patterns across cohorts, and describe data governance procedures and data sharing infrastructure (including pooled and federated analyses, to account for policy limitations on sharing individual data). Descriptive information and procedures for requesting and accessing data are available via the BCRPP data platform (https://epidataplatforms.cancer.gov/bcrpp/). Although originally assembled for risk model development, these data on a large and diverse sample of women with extensive harmonized risk factors and covariates can be a resource to address other outstanding questions in breast cancer epidemiology across time, the lifecourse, and diverse populations. Citation Format: Peter Kraft, Laura Beane-Freeman, Julie Palmer, Ellen O'Meara, Jeanine Genkinger, James Lacey, Thomas Rohan, Lauren Teras, Marian L. Neuhouser, Kala Visvanathan, Celine Vachon, Roger Milne, Christopher Haiman, Yu Chen, Heather Eliassen, Cari Kitahara, Katie O'Brien, Emily White, Garnet Anderson, I-Min Lee, Archie Campbell, Renee Fortner, Ylva Lagerros, Sven Sandin, Mia Gaudet, Montserrat Garcia-Closas, Nilanjan Chatterjee. The Breast Cancer Risk Prediction Project: a resource of 1.8 million women with diverse backgrounds from 20 prospective cohorts to improve breast cancer risk prediction and advance breast cancer epidemiology [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 3583.

Background: Multicellular programs in the tumor microenvironment (TME) drive cancer pathogenesis and response to therapy but remain challenging to identify and profile clinically. Here, we present Spatial EcoTyper, a multimodal machine learning framework for large-scale profiling of spatially dependent cell states and multicellular ecosystems, termed spatial ecotypes (SEs). Methods and Findings: By integrating five million single-cell and spot-level spatial transcriptomes from 10 diverse human neoplasms, including carcinomas and melanomas, we identified a rich tapestry of conserved regional plasticity across nine TME cell types. We then applied Spatial EcoTyper to deeply dissect these data, revealing nine novel SEs with broad conservation - each with unique biology, TME cell states, geospatial features, and clinical outcome associations. SEs were linked to specific biogeographic domains, including the tumor core, margin, and adjacent stroma, and were remarkably well-validated in 235 held-out tumor samples profiled by either single-cell RNA sequencing, sequencing-based spatial transcriptomics (10x Visium), or high-resolution imaging-based spatial transcriptomics (Vizgen MERSCOPE and NanoString CosMx). By deconvolving SEs from over 1,200 bulk tumor RNA-seq profiles, we discovered striking associations between SE levels and response to immune checkpoint inhibitors (ICIs), outperforming nearly 40 potential correlates of ICI response, including previously described signatures of tertiary lymphoid structures and tumor tissue PD-L1 expression. Notably, baseline levels of a proinflammatory SE - localized to the tumor core and defined by a hub of unique T cell, macrophage, fibroblast, and endothelial cell states - strongly forecasted ICI benefit, while another SE - localized to the tumor margin and defined by myofibroblast and hypoxia-associated endothelial cell states - portended ICI resistance. Conclusion: Our data reveal fundamental units of TME organization and demonstrate a platform for large-scale profiling of spatial cellular ecosystems in any tissue, with implications for improved forecasting of immunotherapy response and the development of novel therapeutic strategies. Citation Format: Wubing Zhang,Erin L. Brown,Abul Usmani,Noah Earland,Hyun Soo Jeon,Irfan Alahi,Nicholas P. Semenkovich,Janella C. Schwab,Chloé B. Steen,Chloe M. Sachs,Bilge Gungoren,Faridi Qaium,Peter K. Harris,Jacqueline Mudd,Qingyuan Cai,Antonella Bacchiocchi,Ryan C. Fields,Mario Sznol,Ruth Halaban,David Y. Chen,Aadel A. Chaudhuri,Aaron M. Newman. High-resolution profiling of the tumor microenvironment with spatial ecotypes [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 153.

Abstract Introduction: Fibroadenoma is the most common type of benign breast lesion and is not traditionally considered an indicator of increased breast cancer risk. This study aimed to identify biomarkers within fibroadenoma associated with subsequent breast cancer development using patient samples from the Mayo Clinic Benign Breast Disease (BBD) cohort. Methods: We analyzed formalin-fixed paraffin-embedded (FFPE) breast biopsy samples from patients diagnosed with non-proliferative fibroadenoma who either developed breast cancer later (n=10) or remained cancer-free (n=5). NanoString RNA profiling identified differentially expressed genes in fibroblasts associated with subsequent cancer development. Immunohistochemistry was first done to compare overall expression of PI16 in this cohort. Opal multiplex immunofluorescence staining (mIF) was then performed to evaluate the expression of PI16 in fibroblasts (Vim+PanCK-) and assess immune cell infiltration (CD45+) in different tissue compartments. Fibroblasts were identified by vimentin positivity (Vim+), and epithelial cells were marked by pan-cytokeratin positivity (PanCK+). Results: Compared to patients who remained cancer-free, those who developed breast cancer showed significantly increased expression of PI16 in fibroblasts (PI16+Vim+PanCK-, 3.25-fold increase, p=0.0011). Additionally, the number of immune cells (CD45+) was significantly decreased (0.52-fold, p=0.03) in the fibroadenoma lesions of patients who later developed cancer. Conclusions: Our findings suggest that elevated expression of PI16 in fibroblasts within fibroadenoma may serve as potential biomarkers for predicting breast cancer risk for women with fibroadenoma. PI16 encodes Peptidase Inhibitor 16, which is implicated in modulating fibroblast activity and extracellular matrix remodeling—processes that can contribute to a tumor-supportive microenvironment. Elevated levels of this protein in fibroblasts suggest an altered microenvironment that may foster tumor initiation and progression. The decreased immune cell infiltration further implies an immunosuppressive microenvironment that may facilitate cancer progression. As these markers can be detected in FFPE samples, our results could lead to the development of a readily applicable clinical test to identify fibroadenoma patients at increased risk for breast cancer, enabling earlier intervention and surveillance. Citation Format: Pria Westerman, Nicole Cruz-Reyes, Camila Pacheco Erak, Robert Vierkant, Bryan McCauley, Stacey Winham, Mark Sherman, Amy Degnim, Derek Radisky. Elevated PI16 expression in fibroadenoma fibroblasts is associated with subsequent breast cancer development. [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 2077.

Abstract SWI/SNF chromatin remodeling complexes are frequently mutated across human cancers, including cholangiocarcinoma (CCA). Dysregulation of this complex due to loss of function mutations in AT-rich interaction domain-1A (ARID1A) and polybromo-1 (PBRM1) occur in up to 18% and 10% of cases respectively. SWI/SNF subunit mutations frequently co-occur alongside PI3K/AKT pathway activation. Preexisting literature suggests altered gene expression and protein synthesis contribute to CCA oncogenesis, but our understanding of these mechanisms remains immature. Genetically accurate preclinical models are needed to define the roles of ARID1A and PBRM1 in CCA and identify therapeutic targets. We developed novel murine models of CCA with either ARID1A, PBRM1 or dual ARID1A/PBRM1 (APAC) knockout in the context of AKT activation, utilizing a surgical biliary transfection method for somatic gene editing with CRISPR/Cas9. Within 8 months, mice formed numerous liver tumors, predominantly CCA and mixed CCA/HCC. Knockout of ARID1A and PBRM1 was confirmed through genomic PCR, qRT-PCR, and immunoblot assays. Histologically, murine APAC tumors displayed phenotypic similarities to human CCA, including dense cancer-associated fibroblast infiltration and variable CD45+ immune cell infiltration. A novel, syngeneic CCA cell line was isolated from a dual ARID1A/PBRM1 knockout tumor (APAC-2). RNA-sequencing of primary tumors and the APAC-2 cell line versus a non-invasive cholangiocyte cell line with AKT activation only (AC), was performed to identify differentially expressed genes (DEGs) of interest in ARID1A and PBRM1-deficient CCA. Periostin (Postn) was among the top upregulated genes in the APAC-2 cell line. POSTN is a secreted glycoprotein that is upregulated in human CCA and may be therapeutically targetable. Knockdown of periostin inhibited the in vitro growth of APAC-2 cells and a human CCA cell line with ARID1A mutation and periostin overexpression (HuH-28). These studies serve to characterize and validate a genetically relevant murine model of epigenetic dysregulation in ARID1A and PBRM1-deficient CCA. This model can be used for identification and validation of potential drug targets. Ongoing work focuses on studying multiple dysregulated pathways and testing dependence on genes of interest, including periostin, as potential treatment strategies for CCA. Citation Format: Jayla Millender, Jennifer Tomlinson, Ryan Watkins, Danielle Carlson, Jack Sample, Nathan Werneburg, Rory Smoot, Keith Robertson, Gregory Gores, Caitlin Conboy. Novel murine models of epigenetic dysregulation in SWI/SNF-mutated cholangiocarcinoma [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 2741.

Abstract Tumor associated macrophages are known to fuse with cancer cells resulting in the formation of a cellular phenomenon called tumor-macrophage fusion cells (TMFCs). TMFCs are a highly aggressive cell subtype that may move into circulation, evade immune surveillance, and possibly aid in seeding of metastatic sites. The formation of TMFCs involves CD47, a ubiquitous surface receptor that acts as a “don’t eat me” signal, interacting with the macrophage protein SIRPa, a down regulator of phagocytosis, and syncytin-1, a mediator of hyperploidy cancer fusion. In this analysis, we isolated circulating TMFCs from the blood of n=39 metastatic cancer patients (pts) to evaluate the expressions of CD47, SIRPa and syncytin-1 as it relates to stable or progressive disease via 30 day imaging by PET scan. We conducted a prospective pilot study of n=39 pts with pathologically confirmed metastatic breast (n=17), prostate (n=7), pancreas (n=7), or colon (n=8) cancer receiving standard of care systemic therapy. We isolated TMFCs from 7.5ml peripheral blood using a CellSieve microfilter, and stained with CD47, SIRPa, & syncytin-1. TMFCs were identified based on large cell size (>30 µm) & multinucleated structure, and were categorized into 4 distinct subtypes: A) >2 distinct non-contacting nuclei within a TMFC, B) 2 distinct non-contacting nuclei, C) 2 contacting nuclei, D) a single multinucleated nuclei. CD47, SIRPa, and syncytin-1 expressions were categorized as negative, low, or high. TMFCs subtypes and marker expressions were compared to pts’ clinical status (i.e. progressive, stable or responding) at first available PET/CT scans after blood draw (∼45 days). TMFCs were identified in 72% of pts (n=28/39), average 8 cells/sample, and ranged 1-77 cells. Of the TMFCs, 12% were subtype A, 11% as B, 11% as C, and 66% as D. CD47 expression was seen in 100% of TMFs, with low expression in 39% of subtype A cells, 58% subtype B, 39% subtype C, & 62% subtype D. Notably, CD47 was significantly lower (p=0.010) in subtype B than other groups. Further, the expression of syncytin-1 in subtype B cells was significantly lower (p=0.02) than in other subtypes, and subtype D cells had 75% higher syncytin-1 localized to the nuclear regions. No correlations were found in the SIRPa expression and subtypes. Interestingly, progressing pts had 200% more TMFCs, more subtype B (83%), and significantly lower (p<0.001) CD47 expressing TMFCs than non-progressing pts. In this pilot study on the expression profiles of circulating TMFCs, it appears that specific subtypes of TMFC with low CD47 expression may be associated with progressive disease. Interestingly, this contrasts existing studies on primary tumors which often correlate higher CD47 with more aggressive cancers. This finding suggests that low CD47 cancer cells circulate blood in some pts with progressing disease and may be pertinent to efforts targeting CD47 in cancer. Citation Format: Sonia Muthuraj, Massimo Cristofanilli, Saranya Chumsri, Susan Tsai, Raymond C. Bergan, Cha-Mei Tang, Daniel L. Adams. Low CD47 expression in specific populations of tumor macrophage fusion cells in blood correlates with progressive disease in metastatic cancers [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 1979.

Abstract Deep learning has advanced digital pathology and multi-omics data analysis. However, there is still a critical need to integrate these modalities to better understand complex diseases. Deep learning models, such as Prov-Gigapath, a state-of-the-art pathology foundation model, have been developed for analyzing whole-slide images (WSI). Our previously published deep-learning framework, OmicsFootPrint (OFP), transforms high-dimensional multi-omics data into intuitive genomically-arranged circular images and utilizes transfer learning to overcome the challenges of high-dimensional, low-sample-size data. In this study, we aimed to combine WSI and multi-omics data to develop a framework capable of distinguishing between invasive ductal carcinoma (IDC) and invasive lobular carcinoma (ILC), which is clinically significant as it may help explain differential responses to therapies, such as tamoxifen versus letrozole. We enhanced the performance of our OFP model by dividing each multi-omics circular image (1024x1024x3 pixels) into 16 equal-sized patches (256x256x3 pixels). Each patch was trained using a separate EfficientNetV2 architecture with pre-trained ImageNet weights. The probability outcomes from these 16 models were subsequently combined through a meta-learner model using the AutoGluon framework. Concurrently, WSI images corresponding to the same patients were independently trained, maintaining consistent train/validation/test splits. The Prov-Gigapath model, which was trained using a dataset from the Providence Health System, was employed to extract patch-level embeddings from the WSI data. These embeddings were fine-tuned using a LongNet-based masked autoencoder to predict outcomes. Finally, predictions from the OFP model and the WSI-based Prov-Gigapath model were integrated through a meta-learner, preserving the original train/validation/test split. Model performances (AUC) were evaluated using the held-out test set. The gold standard used for the analysis was histological identification of IDC versus ILC, against which the performance of omics data alone, WSI data alone, and the multi-modal integration approach were compared. We evaluated our integrated model as a proof-of-concept using The Cancer Genome Atlas (TCGA) breast cancer cohort, comprising 92 ILC and 391 IDC samples. Multi-omics data, including gene expression, microRNA, copy number variation, and RPPA, were analyzed using the OFP framework, while WSI data were processed using the Prov-Gigapath method. The results demonstrated that the multi-modal integration approach significantly outperformed individual modalities, achieving an AUC of 0.98 compared to 0.87 for OFP (original), 0.95 for OFP (16-patches), and 0.92 for WSI. Integrating WSI and multi-omics data improves predictive accuracy, significantly improving the classification of breast cancer subtypes over histology alone. Citation Format: Naresh Prodduturi, Xiaojia Tang, Hamid R. Tizhoosh, Kevin J. Thompson, Richard Weinshilboum, Karthik V. Giridhar, Judy C. Boughey, Eric W. Klee, Liewei Wang, Matthew P. Goetz, Vera Suman, Krishna R. Kalari. A multi-modal learning framework to integrate digital pathology image and multi-omics data in breast cancer [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 2429.

The global burden of hepatocellular carcinoma (HCC) continues to rise, driven by increasing metabolic risk factors such as nonalcoholic fatty liver disease, obesity, and type II diabetes. Patients with advanced HCC have limited therapeutic options and a dismal prognosis. While recent advancements in combination therapies—including immune checkpoint blockade (ICB) with anti-PD-1/PD-L1 and anti-CTLA-4 agents and anti-angiogenics—have improved survival for some, up to 80% of patients fail to respond, underscoring an unmet need for new strategies. Emerging data indicate a strong association between increased tumor-infiltrating lymphocytes (TILs) and better responses in ICB-treated HCC, suggesting TIL enhancement as a promising approach. However, HCC’s inherent low immunogenicity and immunosuppressive tumor microenvironment (TME) limit the efficacy of current TIL-promoting strategies. Here, we demonstrate that low-dose intratumoral immunotherapy using the trivalent MMR vaccine—composed of live attenuated measles, mumps, and rubella viruses—modulates TME immune infiltration, increases cytotoxic T cell frequency, and delays tumor growth in multiple murine HCC models. Importantly, intratumoral or systemic MMR administration in combination with dual ICB (anti-mPD-1/anti-mCTLA-4) significantly prolongs survival, promoting tumor-specific and memory immune responses in both subcutaneous and orthotopic HCC models. These findings reveal a unique mechanism of vaccine-induced anti-tumor activity, highlighting MMR as a viable, globally accessible option for enhancing immune responses in advanced HCC therapy. Citation Format: Mulu Z. Tesfay,Aleksandra Cios,Khandoker U. Randal S. Shelton,Bahaa Mustafa,Camila C. Simoes,Steven Post,Thomas Kelly,Moaven Moaven,Lewis R. Roberts,Mitesh J. Borad,Martin J. Cannon,Alexei Basnakian,Nagalo Bolni. Live attenuated MMR vaccines modulate tumor immune cell infiltration and synergize with standard of care to prolong survival in preclinical HCC models [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 945.

Abstract Background: Breast cancer (BC) remains a leading health concern worldwide and is often undetected until it metastasizes, limiting treatment options and worsening prognosis. Early detection of molecular and immune markers in benign breast disease (BBD) could identify women at elevated risk of developing invasive, node-positive BC, enabling targeted surveillance and prevention efforts. Building on this premise, we investigated the immune and molecular landscape of BBD biopsies preceding BC development to enhance risk stratification and inform preventive strategies. Methods: Utilizing the Mayo Clinic BBD cohort, we conducted a case-control study comparing women who developed node-positive BC following their BBD biopsy (cases; n=8) to those who remained cancer-free (controls; n=11), matched on patient age and biopsy date. Differential gene expression was assessed using the NanoString nCounter system to identify differentially expressed genes between cases and controls. We developed a multiplex immunofluorescence (MxIF) approach using the Opal system by Akoya Biosciences for comprehensive marker detection in formalin-fixed, paraffin-embedded (FFPE) tissue samples. Digitized images were segmented for cell phenotyping, and spatial relationships of immune cells were analyzed and compared between cases and controls. Results: Our findings revealed significantly increased expression levels of immunosuppressive genes in the case group, including CD48, CD84, IL-32, and LAIR1. Immunohistochemical analyses showed that controls exhibited higher frequencies of CD68+ (macrophages), CD11c+ (M1 macrophages), and CD206+ (M2 macrophages) cells compared to cases. Notably, controls had a significant increase in CD68+CD11c+ cells, indicative of a pro-inflammatory, antitumorigenic M1 macrophage polarization. Furthermore, our MxIF analyses provided novel insights into spatial biomarker localization, enhancing prognostic accuracy. Conclusions: The decreased numbers of immune cells and increased expression of immunosuppressive genes in BBD biopsies may be associated with an elevated risk of developing axillary node-positive BC. These findings emphasize the importance of dissecting molecular and immune interactions in BBD for risk assessment. Our results have the potential to improve individualized risk stratification, leading to more targeted surveillance and informed screening. This research not only contributes to our understanding of BC pathogenesis but also opens avenues for developing novel preventive strategies, potentially reducing BC incidence and mortality through early intervention. Citation Format: Matilde Rossi, Nicole Cruz-Reyes, Trista Navarro, Melody L. Stallings-Mann, Bryan M. McCauley, Robert A. Vierkant, Stacey J. Winham, Mark E. Sherman, Amy C. Degnim, Derek C. Radisky. Immune and molecular markers in benign breast disease predict risk of node-positive breast cancer [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 1912.

Genome-wide association studies (GWAS) of multiple myeloma (MM) in individuals of genetic European ancestry (EA) have identified 35 susceptibility loci. Co-heritability analyses have revealed a strong genetic correlation between MM and its precursor condition, monoclonal gammopathy of undetermined significance (MGUS). Previous research validated the association between a polygenic risk score (PRS) based on 23 MM risk loci and the risk of MGUS. Went and colleagues recently reported 12 new MM risk loci. We constructed a MM-PRS using all 35 MM risk variants and assessed its association with MGUS. Additionally, among individuals with MGUS, we assessed associations of MM-PRS with established prognostic factors [M-protein (<0.2 g/dL; 0.2-1.5 g/dL; >1.5 g/dL); isotype (IgG, IgA, IgM); free light chain (FLC) ratio (Normal (0.26-1.65), Abnormal (<0.26 or >1.65))]. Our study population included 1,723 individuals with MGUS seen at Mayo Clinic, Rochester, MN and 14,175 controls from the Mayo Clinic Biobank who lived in the 27-county region around Rochester, MN, and screened negative for MGUS using MALDI-TOF MS. The majority of MGUS patients and controls were of EA. We used logistic or multinomial regression to calculate odds ratios (ORs) and 95% confidence intervals (CIs). All models were adjusted for age, sex, and the first three principal components. The MM-PRS was significantly associated with MGUS when assessed continuously (OR, 1.28 per standard deviation (SD); 95% CI, 1.22-1.35). When comparing to the middle quintile of the MM-PRS distribution, individuals in the highest quintile had 1.41-fold increased risk of MGUS (95% CI, 1.21-1.64), and individuals in the lowest quintile had 0.68 decreased risk of MGUS (95% CI, 0.57-0.81). Of the 35 SNPs evaluated, 27 demonstrated consistent directions of effect for risk of MGUS with published risk estimates for MM. The effect of MM-PRS differed by isotype (P = 2.2 x 105), with higher MM-PRS associated with IgA MGUS (OR, 1.40 per SD; 95% CI, 1.08-1.81) and a lower MM-PRS associated with IgM (OR, 0.66 per SD; 95% CI, 0.52-0.84) compared to IgG MGUS. The MM-PRS was higher in MGUS patients with M-protein > 1.5 g/dL compared to M-protein of size 0.2-1.5 g/dL, although not statistically significant (OR, 1.41 per SD; 95% CI, 0.95-2.08); there were no differences in the MM-PRS between MGUS patients with M-protein size <0.2 g/dL compared to 0.2-1.5 g/dL (OR, 1.01 per SD;95% CI, 0.81-1.26). There were also no differences in MM-PRS by FLC ratio, (OR, 0.95 per SD; 95% CI, 0.852-1.10 for abnormal vs. normal). Our findings of an association of the MM-PRS with risk of MGUS provides further evidence for shared heritability between MGUS and MM. Future studies are needed to examine whether the expanded MM-PRS adds to existing models for MGUS progression. Citation Format: Angelica Macauda, Alyssa Clay-Gilmour, Cristine Allmer, Danelle Moonen, Aaron D. Normann, Nicholas J. Boddicker, Joselle M. Cook, Linda Baughn, Elizabeth E. Brown, Vincent S. Rajkumar, Esteban Braggio, David L. Murray, Susan L. Slager, Shaji Kumar, Celine M. Vachon. Implications of multiple myeloma polygenic risk scores (PRS) for MGUS [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 2274.