Guzman, Emmanuel Contreras

Abstract Germinal center (GC)-derived lymphomas arise within a specialized immune ecosystem, yet whether malignant cells direct its remodeling remains unclear. Integrating genomic, transcriptomic, spatial, and functional analyses across follicular lymphoma and diffuse large B-cell lymphoma, we define a developmental framework linking malignant B-cell differentiation state to tumor microenvironment (TME) organization. Rather than segregating by histology, lymphomas align along a shared differentiation continuum in which proliferative dark zone (DZ) states associate with immune-depleted TMEs, whereas post-GC/memory B-cell (MBC) states drive inflamed but immunosuppressed TMEs. Spatial profiling reveals collapse of GC architecture and emergence of suppressive myeloid niches along this trajectory. Single-cell lineage reconstruction uncovers a DZ-to-MBC developmental axis underlying lymphoma evolution. Genetic reprogramming in vivo demonstrates that malignant differentiation state is sufficient to instruct TME remodeling. Collectively, these findings establish malignant differentiation state as a causal determinant of immune architecture across GC-derived lymphomas and identify lineage-directed immune reprogramming as a potential therapeutic strategy.

Abstract Large Language Models (LLM) are being widely adopted into the medical field for their impressive ability to analyze and summarize large amounts of text data. These models enable clinicians and researchers to extract meaningful insights from complex datasets and may assist with decision making. Here we present our workflows and application of LLM for the interpretation and summarization of clinical data related to CAR-T cell therapy. Using an LLM (Gemini 2.5 pro) in the Google Cloud Computing (GCP) environment, two applications were developed to analyze CAR-T cell therapy clinical data: 1) extracting and summarizing CRS and ICANS event-related data to streamline the compliance team workflow, and 2) identifying features available at time of CAR-T infusion able to classify patients into high- or low-monitoring needs 14 days post CAR-T. Patient data (vitals, labs, hematology notes, and EKGs) were extracted from the electronic medical record (EMR) using Google BigQuery into SQL tables in GCP. For each application, relevant data fields were retrieved, formatted into JSON objects, and embedded in the LLM prompt for context-aware processing. Both applications have undergone iterative prompt engineering after analyzing the LLM output against ground truth data in the EMR. For the application extracting and summarizing CRS and ICANS events, the LLM was optimized on Mayo Clinic Rochester data. When compared to the IEC compliance database, the LLM achieved 100% accuracy and F1 score for CRS events, and 96% accuracy and 82% F1 score for ICANS events. The match rate for CRS and ICANS grades were 83% and 89% respectively. We applied the same LLM to Mayo Clinic Arizona (MCA) and Mayo Clinic Florida (MCF) cases, who document clinical notes and toxicity flowsheet differently, and achieved an accuracy (MCA: 93%, MCF:93%) and F1 score (MCA: 96%, MCF: 96%) for CRS and accuracy (MCA: 81%, MCF:82%) and F1 score (MCA: 76%, MCF: 75%) for ICANS. LLM was able to capture events missed by manual reviews. For most of the discrepancies where compliance team final adjudication is needed, LLM will be updated to flag discrepancies for review by the compliance team. For the application related to the monitoring needs 14 days post CAR-T, the LLM identified 5 categories predictive of high or low monitoring needs post CAR-T infusion (disease status, inflammatory and tumor burden markers, hematologic status, renal function and performance status). Our model achieved a sensitivity of 85.7%, specificity of 23.8%, and F1 score of 65.5% for our first cohort, compared with data extracted by the LLM from the EMR. For our second cohort, with demographics statistically similar to cohort 1 and using the same 5 categories, the LLM achieved a sensitivity of 83.3%, a specificity of 28.6%, and F1 score of 65.4%. Our workflow and applications of LLM’s provide examples and guidance to others interested in applying LLM’s for clinical and research applications. Citation Format: Emmanuel Contreras Guzman, Matthew Jankowski, Andre De Menezes Silva Corraes, Malvika Gupta, Monica L. Shaw, Madiha Iqbal, Talal Hilal, Saurabh Chhabra, Ricardo Daniel Parrondo, Jody K. Mclean, Kim R. Riester, Kayla Joseph, Melinda Tan, Holly Ross, Cleyonia Barnett, Sylvia Carter, Semy Girmay, Rachel Wolan, Milana Ramsey, Christian Downhour, Kristy Morgan, Shae Sibley, Erica Rushing, Lucy Holmes, Allison Burgstahler, Stephen M. Ansell, Hassan Alkhateeb, Matthew Hathcock, Ramona Bruno, Allison C. Rosenthal, Hemant Murthy, Patrick B. Johnston, Jonas Paludo, Yi Lin. Applications of large language models to CAR-T cell therapy clinical data using Google cloud computing [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 2740.

Abstract Introduction: CAR-T therapy has transformed cancer therapy. with its first Food and Drug Administration (FDA) approved product in August 2017. Given the expanding indications and growing experience, particularly with management of immediate toxicities requiring specialized expertise such as Cytokine Release Syndrome (CRS) and Immune Effector Cell-Associated Neurotoxicity syndrome (ICANS), FDA recently removed the Risk Evaluation and Mitigation Strategies (REMS) for CD19 and BCMA targeting CAR-T on June 26th2025. In particular, given that CRS and ICANS typically occur in the first 2 weeks post CAR-T infusion, the minimum required stay near CAR-T treatment center is reduced from one month to 2 weeks. While this could reduce treatment burden for patients (pts) who need to travel away from home to receive CAR-T, there is a need to update the transition of care from CAR-T center to pts' local oncology practice. Pts carry in their rate of recovery and have varying supportive care needs in the second half of the month post CAR-T infusion. To assist with care transition and planning, we examine the use of AI multimodal large language model (LLM) to predict patients' supportive care needs 2 weeks after CAR-T infusion based on the clinical data at the time of CAR-T evaluation, prior to leukapheresis. This timepoint was selected to also maximize travel, lodging and caregiver planning for pts and family as well. Methodology: Pts who were treated with FDA-approved CAR-T at Mayo Clinic Rochester between December 2022 and June 2025 were examined. Based on experiences with referral practices in the Midwest region, we defined patients with high monitoring needs as those who between day 14-30 post CAR-T had either hospital admission due to recurrent CRS or ICANS or required transfusion of more than 1 blood product per week (red blood cell or platelets). Others were classified as low monitoring needs. Clinical notes, labs, vital signs and electrocardiogram at the time of CAR-T evaluation were abstracted and summarized by Gemini LLM for association with patients with high versus low monitoring needs post CAR-T. Model was developed and tested for predicted needs and concordance with actual outcome of these pts. Results: Among the 362 pts who received FDA approved CAR-T in the timeframe, 84 pts had high monitoring needs. An additional 84 pts with low monitoring needs were randomly selected from 278 pts. These were divided into a training cohort and test cohort of 84 pts each, with 42 pts with high and 42 pts with low monitoring needs. Demographics of the two cohorts displayed no statistical difference. LLM analysis of the clinical data at the time of CAR-T eval identified 5 categories associated with risk for high monitoring needs: 1) Disease status (clinical notes about bulky, progressive disease or CNS involvement); 2) inflammatory and tumor burden markers (CRP, LDH); 3) hematologic status (platelet, plt; neutrophil count ANC); 4) renal function; 5) performance status. Using common clinical lab parameter convention (platelets <100, ANC< 1, CrCl< 30ml/min), and prioritizing for not missing pts with high monitoring needs, LLM model was defined to predict patients for high monitoring needs if high risk feature was identified in category 1 or any category combinations; whereas patients with no high risk feature or only 1 category that is not disease status was predicted as low monitoring needs. In this model, the sensitivity was 85.7%, specificity 23.8%, F1 score 65.5%. Applying this model to test cohort 2, the sensitivity was 83.3%, specificity 28.6%, and F1 score 65.4%. Discussion: We demonstrate proof-of-concept that LLM can be used to analyze clinical data at the time of evaluation for CAR-T eligibility to predict for supportive care needs post treatment. Not surprisingly, features identified by LLM to be associated with high monitoring needs are those previously reported by us and others at the time of CAR-T evaluation and or pre-lymphodepletion to be associated with risk for cytopenia, infection, hospitalization and non-relapse mortality. Based on the summary output generated by LLM, additional opportunities for classification methods were identified to improve specificity and balanced accuracy of the LLM model. This is a promising tool for further development to assist clinicians and patients to plan their treatment journey.

Abstract Introduction: Suboptimal response to therapy, early disease relapse, and limited long-term survival remain critical challenges in a substantial number of patients with multiple myeloma (MM). While cross-sectional imaging studies are commonly obtained in clinical practice, only limited information is currently incorporated into risk stratification efforts and clinical decision making. Here, we investigate the use of novel image-based biomarkers to identify patients at high risk for adverse outcomes. Methods: We obtained the pretreatment 18F-FDG PET/CT scans of patients with newly diagnosed MM with available clinical and laboratory data seen at Mayo Clinic between 2006 and 2022. In this analysis, we included 637 patients with detectable metabolic tumor volume on their pretreatment scan. The native DICOM images were pre-processed and subjected to automated image segmentation using nnU-Net (nnU-Net v1.0, inference using a pre-trained hematologic malignancies model). Limitations of this automated segmentation approach include heterogeneity introduced by varying image quality, technical variability in PET reconstruction, and potential inclusion of non-neoplastic FDG-avid lesions in segmented regions. PyRadiomics was used to extract quantitative radiomics features for clinical prediction modeling. We evaluated the predictive power of tumor size-, shape-, and texture-related radiomics features for identifying high-risk disease (advanced ISS stage at diagnosis, the presence of high-risk FISH abnormalities, failure to achieve a stringent complete response [sCR] to first-line therapy by IWMG criteria, and early mortality). All radiomics features were quantile-normalized for downstream modeling. Forward and backward feature selection as well as testing for multicollinearity were used to remove redundant features. Results: Median age at diagnosis of the 637 myeloma patients was 64 years (range 22–90), and 387 patients (61%) were male. After a median follow-up of 4.4 years (95% CI 0.2–11.7), median overall survival (OS) was 7.3 years (95% CI 6.3–8.3). Increased total tumor volume (upper tertile vs lower tertiles), measured by the total tumor surface area, was associated with advanced ISS stage (III vs I+II; OR 1.73, 95% CI 1.24–2.42, p=0.001, n=637). Increased tumor density (dichotomized at the median), measured by the median gray level intensity (OR 1.93, 95% CI 1.94–3.12, p=0.007, n=303), and decreased tumor heterogeneity (lower tertile vs upper tertiles), measured by the variability of gray-level intensities (OR 1.88, 95% CI 1.49–3.12, p=0.012, n=303), were associated with the presence of high-risk cytogenetics. Both increased total tumor volume (4.9% increase per tertile, 95% CI 2.2–7.6, p<0.001, n=600) and tumor density (8.9% increase per tertile, 95% CI 4.5–13.3, p<0.001, n=600) were independently associated with more extensive bone marrow involvement at diagnosis. Increased tumor heterogeneity (upper tertiles vs lower tertile), as measured by rapid changes of gray level intensities, was associated with failure to achieve a stringent complete response to first-line therapy (OR 0.50, 95% CI 0.30–0.82, p=0.006, n=351). Increased tumor volume was associated with shorter OS (HR 1.35 per tertile, 95% CI 1.11–1.63, p=0.002, n=301), independent of age, sex, ISS stage, and the presence of high-risk cytogenetics. A simple additive score consisting of tertiles of total tumor volume, ISS stage, and plasma cell labeling index ≥2% identified patients at risk of mortality within 3 years of diagnosis (OR 2.00, 95% CI 1.40–2.85, p<0.001, n=126), with moderate discrimination (AUC=0.71). Sensitivity analysis demonstrated total tumor volume and texture as most predictive of overall survival in a random forest survival model (C-index 0.712, 95% CI 0.649–0.775) and tumor texture as most predictive of response to therapy in a random forest classification model (C-index 0.693, 95% CI 0.607–0.779). Conclusions: Automated image segmentation and radiomics feature extraction from 18F-FDG PET/CT scans produces biomarkers that capture disease stage, burden, and risk in newly diagnosed MM. Total tumor volume and heterogeneity are strongly associated with important disease characteristics and can be used to identify patients at high risk for therapeutic resistance, early mortality, and inferior long-term outcomes.

Abstract Despite a characteristic indolent course, a substantial subset of follicular lymphoma (FL) patients has an early relapse with a poor outcome. Thus far, efforts to identify factors that predict survival have been unsuccessful. However, we and others have demonstrated the prognostic relevance of CD4+ T cells in the tumor microenvironment (TME) of FL and developed a bio-clinical risk model (called BioFLIPI) that offers improved risk stratification (Mondello et al, BCJ 2021). However, the mechanisms defining different BioFLIPI scores remain to be fully understood. To identify the programs associated with the BioFLIPI scores, we first interrogated the RNA-seq of purified B cells from 67 newly diagnosed FL1-3A (BioFLIPI 1 n=21; BioFLIPI 2-3 n=36; BioFLIPI 4 n=10). We found 1,156 upregulated genes and downregulation of 930 genes in BioFLIPI 4 vs 1, including well-known immune molecules such as CD70, IL10RA, CD69, and CCR6 (FDR<0.05). Gene set enrichment analysis (GSEA) using the MSigDB confirmed a negative enrichment in multiple immune response and interferon gamma signatures and revealed striking positive enrichment for cell proliferation and antigen presentation pathways in BioFLIPI 4 compared to BioFLIPI 1 tumors. The same pathways were dysregulated in BioFLIPI 2-3 but a lower magnitude. To investigate the somatic mutations associated with different BioFLIPI scores, we performed WES of the same 67 FLs and further expanded this cohort for a total of 90 patients (BioFLIPI 1 n=26; BioFLIPI 2-3 n=51; BioFLIPI 4 n=13). We found enrichment for mutations in several regulators of the germinal center in BioFLIPI 4 and 2-3, such as MEF2B (p=0.029), FOXO1 (p=0.038) and IRF8 (p=0.035), suggesting the cooperation of multiple oncogenic events promoting lymphomagenesis. In contrast, BioFLIPI 1 tumors harbored a higher rate of mutations in TNFRSF14 gene, known to enhance anti-lymphoma immune response. To investigate whether BioFLIPI modulates immune signatures of the microenvironment, we performed TME deconvolution by applying CIBERSORTx to bulk RNA-seq of these 67 FLs. We found that FLs with BioFLIPI 4 and 2-3 were enriched for CD4+ Tcm exh (p=0.02), Tem exh (p=0.04), Tfhexh (p=0.07) and NK cells (p=0.01) compared to BioFLIPI 1. In line with this data, CyTOF performed on a subset of these patients (BioFLIPI 1 n=12; BioFLIPI 2-3 n=25; BioFLIPI 4 n=6) confirmed progressive increase of CD4+ Tcm exh (p=0.006), Tem exh (p=0.009) and Tfh exh (p=0.025) cells from BioFLIPI 1 to BioFLIPI 4. Single cell RNA/BCR-seq of 21 FLs (BioFLIPI 1 n=6; BioFLIPI 2-3 n=12; BioFLIPI 4 n=3) corroborated a higher abundance of Tem exh, cytotoxic (CTL) Tem exh, Tfhexh and NK cells in BioFLIPI 4 and 2-3. Conversely, CD8+ naïve T cells were prevalent in BioFLIPI 1 tumors. Differential gene expression in T cell clusters from BioFLIPI 4 vs BioFLIPI 1 FL revealed a marked upregulation of multiple immune genes involved in antigen presentation (e.g., HLA-DRB, HLA-DQA, HLA-DPA/B) and immune response (e.g., ETS1, TNFSF8, IL7R). GSEA showed a positive enrichment for antigen presentation signatures and negative enrichment for TNFα signaling and Oxphos metabolism. Similar findings were observed in BioFLIPI 2-3 FLs. Using CellChat analysis, we found a stronger interaction between malignant B cells and Tem exh and CTL Tem exh cells in BioFLIPI 4 and 2-3 vs BioFLIPI 1 tumors. This was linked to an increased strenght of MHCI-CD8 interaction and concordant decrease of MHCII-CD4. Notably, the suppressive BTLA-TNFRSF14 signaling axis was present in BioFLIPI 4 and 2-3 but not in BioFLIPI 1 FLs. This was accompanied by a decreased signaling through TNF/LTA-TNFRSF1B, CD80/CD86-CD28/CTLA4. Linking the BioFLIPI-specific immunogenicity to lymphomagenesis, we found that FL patients with BioFLIPI 4 had two-three folds higher risk of transformation to an aggressive lymphoma (12% vs 7%) or death (21% vs 3%) than those with BioFLIPI 1, suggesting that this immune dysfunction favors the expansion of a dominant tumor clone and promotes a more aggressive form of FL. Collectively, our findings demonstrate that BioFLIPI scores in FL reflect distinct patterns of tumor immunogenicity that shape the TME. Notably, high BioFLIPI scores are paradoxically associated with enhanced antigen presentation alongside a suppressed immune response. This suggests a sustained antigen-specific T cell response that may drive immune exhaustion, promote tumor progression, and ultimately lead to histologic transformation

Abstract Background Chimeric-antigen-receptor (CAR) T-cell therapy has transformed outcomes for relapsed or refractory hematologic malignancies, yet early identification of patients at highest risk for death remains elusive. A prognostic model that leverages only the information routinely available at the time of CAR-T evaluationcould guide pre-infusion counseling, intensity of follow-up, and resource allocation. Methods We analyzed a prospectively maintained cohort of CAR-T recipients treated at Mayo Clinic to develop and test our model. For each patient, data (vital signs, lab results, cardiac assessments [ECG and echocardiography], and PET/CT images) from initial CAR-T evaluation (up to 60 days prior to CAR-T infusion) was considered. Vital signs and labs were left un-imputed, while ECG, echocardiographic parameters, and PET/CT features were forward-filled from the preceding 60 days. Each patient-day was labeled as “outcome present” or “no outcome,” enabling supervised training of gradient-boosting models to predict overall survival (OS) at 6 months post CAR-T infusion. Model performance was evaluated using the area under the ROC curve (AUC) with 10,000-fold bootstrapping with data shuffling between each iteration to derive confidence intervals for stability. To enhance interpretability, we also developed a simplified model using the top 19 features (determined by Shapley value-based importance), aiming to preserve predictive accuracy with a smaller feature set. Results A total of 570 CAR-T cell treated patients were included in this study (476 patients in the training cohort and 94 patients in the validation cohort). The median age was 64 years and 61% males. The primary diagnoses were B-cell lymphoma (324 patients) and multiple myeloma (246 patients). A total of 4,851 individual vital-sign readings, 11,781 laboratory measurements were included. PET/CT images were available in 318 patients. The ML model demonstrated high discriminative ability for predicting OS at 6 months post CAR-T infusion. Vital signs alone produced a mean AUC of 0.63; radiomics features alone produced a mean AUC of 0.71, while laboratory tests alone produced a mean AUC of 0.84. The features most contributory to each single-modality model were then combined into a simplified multimodal model with a total of 19 features, which produced a mean AUC of 0.87, with a 95% CI of 0.73 - 0.97. The 19 features included in the simplified model were: LDH, ferritin, uric acid, eosinophils, fibrinogen, D-dimer, heart rate, mean arterial blood pressure, QTc, QT interval, and the radiomics features of sphericity, first-order Kurtosis, first-order mean, first-order energy, first-order total energy, shape maximum2D diameter (sagittal plane), SUV CoV, SUVp90, shape flatness. This multimodal predictive model was then applied to the validation cohort (uninvolved with model training or testing), which produced an AUC of 0.93, with a 95% CI of 0.84 - 0.98. Conclusions Our ML-driven approach successfully predicts overall survival 6 months post CAR-T infusion, highlighting those individuals at most risk of poor outcomes at the time of CAR-T evaluation. The strong performance of the multimodal version of the model suggests that each data domain contributes complementary resolving power, and by including novel data types, such as PET CT radiomics features, the most accurate predictions could be realized. By providing accurate, near-real-time risk stratification for poor outcomes within 6 months of CAR-T infusion, the model could facilitate proactive patient management. Prospective studies of the integration of this ML model into CAR-T treatment workflows will be crucial to confirm its clinical benefits and to facilitate its adoption in practice.

Abstract Twenty percent of patients with follicular lymphoma relapse early with poor outcomes; however, the molecular mechanisms underlying this aggressive behavior are unknown. Using a multiomic approach, we show that patients with follicular lymphoma with elevated IRF4 expression (IRF4hi) have increased transformation risk, dysregulated immune signaling, and a suppressive tumor microenvironment. Loss- and gain-of-function experiments in IRF4hi lymphoma cells, along with chromatin profiling, demonstrate that IRF4 impairs their interaction with T cells by repressing antigen presentation and co-receptor gene modules while promoting the expression of cytokines that antagonize T-follicular helper cell and regulatory T-cell functions. Additionally, IRF4 rewires tumor metabolism and restricts glucose availability to immune cells. Silencing of IRF4 inhibits tumor cell growth and restores immune surveillance mechanisms, thus representing a promising target for therapy. Our data suggest that IRF4hi lymphoma cells co-opt a developmental mechanism used to exit the germinal center response in promoting a more aggressive cancer via engagement of multiple immune-evasive mechanisms. Significance: We show that IRF4 controls immune and metabolic signaling with functional effects on the cross-talk between B and T cells, resulting in a suppressive immune microenvironment when IRF4 is overexpressed in lymphoma. This underscores the importance of IRF4 as a pivotal therapeutic target to restore antitumor immunosurveillance for IRF4-associated lymphomas.

e13622 Background: Chimeric antigen receptor T-cell (CAR-T) therapy has revolutionized treatment of hematologic malignancies; however, severe toxicities like cytokine release syndrome (CRS) remain a limitation to broader utilization. Using a large dataset of patients treated with CAR-T, we developed a machine learning (ML) model leveraging multi-model patient characteristics to predict the risk of CRS onset with clinically meaningful lead times. Methods: A prospectively maintained dataset of CAR-T patients treated at Mayo Clinic was used. Baseline disease/patients characteristics, labs, ECG, echocardiogram, and vital signs (including data from remote patient monitoring program) were collected. A 24-hour rolling window was used to generate daily averages, maximum, minimum, and deltas for each parameter. ECG and echocardiogram data were forward-imputed; vital signs and labs were not imputed. Each day-post-infusion served as a potential event (CRS onset vs. no CRS). A gradient boosting classifier was trained and tested on a subset of patients (n = 351) using 1,000 bootstrap iterations, evaluating 24-, 48-, and 72-hour lead times. Shapley-based feature importance guided simplification to an 11-feature model, externally validated on an independent cohort (n = 51). Models’ performance assessed using the area under the receiver operating characteristic curve (AUC). Results: A total of 402 patients treated with CAR-T were examined. Median age was 64 years and 36% were female. Primary diagnosis was B-cell lymphoma (n = 240) or multiple myeloma (n = 162). A total of 226,332 vital sign measurements and 372,871 lab values, as well as ECG and echocardiogram data, were collected for 30 days post-infusion. Overall, 79.8% of patients experienced CRS within 30 days. The full ML model demonstrated AUCs of 0.90 (95% CI 0.86–0.93), 0.83 (95% CI 0.77–0.88), and 0.82 (95% CI 0.77–0.87) for 24-, 48-, and 72-hour predictions of CRS, respectively. Including “day of CRS onset post CAR-T” as a feature improved the 72-hour AUC to 0.87 (95% CI 0.83–0.90). A model restricted to vital signs alone showed AUCs of 0.88, 0.75, and 0.71 for 24-, 48-, and 72-hour horizons. An 11-feature model—focusing on temperature, blood pressure, fibrinogen, CRP, serum sodium level, QRS interval, and lymphocyte counts—maintained high accuracy at 24 hours (AUC 0.90, 95% CI 0.86–0.93) and was validated externally using an independent dataset with an AUC of 0.92. Conclusions: Our ML model provides a reliable approach for predicting if/when CRS will occur, with up to three days of lead time before clinical manifestation, offering an opportunity for proactive management. The robust performance of a parsimonious 11-feature model supports practical adoption across clinical settings and real-time risk stratification to enhance patient safety. Prospective validation and integration into clinical workflows will be critical to translating these findings into practice.

Introduction The tumor ecosystem in multiple myeloma (MM) involves complex interactions among various cell types, including clonal plasma cells (PCs) and immune cells. Previous research has indicated that the bone marrow (BM) microenvironment's metabolite and lipid profiles differ between MM patients and those with precursor conditions. This study aims to explore if differences in cell composition, metabolic phenotype of clonal PCs, and the spatial organization of various cell types are linked to the early versus late progression of high-risk smoldering MM (SMM) to MM. Methods We identified high-risk SMM patients using the IMWG 20-2-20 criteria, focusing on early progressors (EP), who progressed to MM within one year, and later progressors (LP), who progressed after more than four years. Using a high-dimensional single-cell spatial imaging platform, CODEX (CO-Detection by indEXing), we identified clonal PCs and immune cell types as well as their metabolic phenotypes at the single-cell level, in paired FFPE BM samples at SMM diagnosis and at the time of progression to MM. Our validated panel included markers such as CD3e, CD4, CD8, CD14, CD20, CD31, CD38, CD45, CD66a, CD68, CD138, HLA-DR, ASCT2, pNRF2, GLUT1, LDHA, ATPA5, CPT1A, and Citrate synthase. We employed single-cell segmentation, artifact cleanup, signal thresholding, and spatial analysis. Repeated measure models accommodated multiple regions of interest (ROIs) per patient and accounted for potential intrapatient correlation, while Poisson models estimated incidence rate ratios (RR) of cell types. Spatial statistical models evaluated cell type segregation and integration at the time of SMM diagnosis. Results Single cell events were acquired from 6 unique SMM patients who had paired FFPE BM samples from the time of their SMM diagnosis and at the time of their progression to MM. There were 3 EP SMM patients (Time to Progression (TTP):8, 8, 10 mos) and 3 LP SMM patients (TTP: 51, 58, 117 mos). The median number of single cell events identified per patient slide sample was 104,390 (Range: 24,357 - 314,877). We identified 8 main cell types: PCs, CD4+ T-cells, CD8+ T-cells, B-cells, granulocytes, Monocyte/Macrophages and two groups of other immune cells. The PCs and the remainder of the non-PCs were further classified based on the expression of the following metabolic markers: ASCT2 (Glutamine importer), CPT1A (Fatty acid metabolism), GLUT1 (Glucose importer), LDHA (Glycolysis), Citrate synthase and ATPA5 (Oxidative Phosphorylation). There was a higher RR of granulocytes (RR 2.1, CI 1.6-2.8, p<0.001) in EP compared to LP samples at the time of SMM diagnosis, but a lower RR of CD4 T cells (RR 0.40, CI 0.3-0.6, p< 0.001). Focusing on metabolic phenotypes of the PCs, we observed higher incidence RR of most metabolic markers: ASCT2 (RR=1.77; p=0.032), ATPA5 (RR=1.25; p<0.001), CPT1a (RR=2.42; p=0.015), and LDHA (RR 2.13; p<0.001) in EP compared to LP samples at the time of SMM diagnosis. However, GLUT1 had a lower incidence RR in EP patients (RR=0.79; p<0.001). The same modeling for non-PCs at the time of SMM diagnosis yielded higher expression of LDHA (RR=1.98; p<0.001) and ASCT2 (RR=3.3; p=0.002). Using the x and y coordinates and the corresponding phenotype classifications, we performed spatial statistical modeling based on spatial point processes to generate aggregated measures of how different types of cells of interest interacted. CD4 T cells were more isolated in LPs compared to EPs (p =0.046). Segregation indices for PCs and CD4 T cells indicated that they were more integrated/collocated in EPs. Granulocytes were more isolated in EPs although the difference was not as pronounced (p= 0.07). PCs were more integrated with granulocytes (p=0.06) and B cells (p=0.07) in LP patients. Further analyses of paired samples at the time of progression to MM will be reported at the time of the presentation. Conclusion This study is the first to provide whole-slide, single-cell, spatial proteomic characterization of clonal PCs and immune cells in the FFPE BMs of high-risk SMM patients with varying progression times to MM. Distinct metabolic pathway markers in clonal PCs and their spatial relationships with immune cell subsets are closely associated with the risk of progression to MM. These findings illuminate the complex interactions between the metabolic states of clonal PCs and their spatial orientation relative to immune cells in the pathogenesis of MM.