Next-generation sequencing has transformed cancer care by providing essential insights for diagnosis, prognosis, and treatment. However, variability in testing timing, reporting practices, and interpretation challenges limits its clinical impact. This article highlights key opportunities to optimize somatic reporting, emphasizing the importance of timely testing throughout the cancer care continuum to maximize the diagnostic and therapeutic relevance of findings. Technical factors such as test design, sequencing depth, and the use of liquid biopsy substantially influence result accuracy and interpretation, underscoring the need for careful integration with clinical history. Standardized reporting practices that clearly delineate diagnostic, prognostic, and therapeutic findings can enhance the clinical utility of next-generation sequencing results. Streamlined formats and curated clinical trial data further support actionable decision making. Additionally, direct patient engagement and education are essential for empowering patients to navigate genomic testing and make informed decisions about their care. By leveraging multidisciplinary tumor boards, decision-support tools, and emerging artificial intelligence technologies, clinicians can better navigate the complexities of somatic reports. Standardization and clarity in reporting are critical to advancing precision oncology, empowering providers and patients to make informed treatment decisions and improve outcomes.

Abstract Background: Studies in cancer early detection have revealed circulating proteins to be powerful and informative biomarkers. We further aimed to develop a machine-learning framework robust to batch effects across independently processed datasets and to identify a subset of proteins capable of detecting cancer at its earliest stages. In this study, we sought to evaluate the early-access Illumina protein prep proteomic assay not only for its technical reproducibility but also for its ability to yield biologically informative protein signatures relevant to cancer early detection. Experimental Procedures: Proteomic profiles from six plates were analyzed, encompassing approximately 217 normal and 206 cancer plasma samples. The first four plates contained samples collected from Eastern European sources and were used for model training and feature selection, while the remaining two plates contained samples collected in the United States from populations with diverse ancestry backgrounds and served as an external test set. These two plates included samples from bladder, breast, gastric, and lung cancers, representing diverse biological and technical conditions. Feature selection was performed using the Minimum Redundancy-Maximum Relevance (MRMR) method, which ranks proteins by maximizing mutual information with cancer status while minimizing redundancy. The top 200 proteins were used to train a Support Vector Machine (SVM) classifier with a radial-basis kernel. Results: The model achieved a mean AUC of 0.89 across the two independent test plates, demonstrating strong cross-batch and cross-ancestry reproducibility and confirming that informative, generalizable protein features can be extracted from the Illumina platform. At 95% specificity, the classifier achieved an overall sensitivity of 65% (95% CI: 51-77%), with particularly strong performance in Stage II cancers at 82% (95% CI: 52-95%), underscoring its potential utility for early detection. Performance was consistent across cancer types, with highest sensitivities observed in lung. Importantly, a six-fold leave-one-plate out cross-validation yielded an average sensitivity of 82% at 99% specificity, demonstrating that integrating diverse data sources will likely strengthen model generalizability. Conclusions: A machine-learning framework applied to large-scale proteomic data identifies a compact and biologically meaningful subset of proteins capable of early cancer detection. The results highlight the robustness of the Illumina Protein Prep, 6K assay, and the feasibility of developing batch-insensitive protein classifiers for population-scale cancer screening. Citation Format: KAMEL LAHOUEL, Mete Mulazimoglu, Kameron Bates, Candice Wike, Kunjur Manasa Upadhyaya, Victoria Zismann, Kamawela Leka, Payton Smith, Gracyn Benck, Kianna Martos Rupp, Chaney Jambor, Matteo Munini, Sophie Pénisson, Stephanie Pond, Jeffrey Trent, Patrick Pirrotte, Cristian Tomasetti. Early cancer detection using early-access Illumina protein prep 6K assay and machine learning [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 7614.

Abstract CML is defined by evolution from chronic phase (CP) to increased disease burden during blastic phase, but the cellular mechanisms that create these disease states and produce the transition between states is not understood. We previously used state-transition models to show that CML evolution is not encoded in single-cell transcriptional microstates but instead emerges only when gene expression is aggregated into population-level macrostates where distinct phenotypic disease states emerge. Here, we extend this framework to ask how antagonistic teams of genes and their regulatory network defined steady states give rise to these macrostates. Using weekly time-series single-cell RNA sequencing from both CP and blast crisis (BC) inducible CML mouse models, we assessed the origin of phenotypic disease macrostates in each cell type by identifying antagonistic teams of genes. We identified these teams for each cell type lineage by selecting the genes where their eigenvalue in the state-space construction and their observed expression change combine to indicate that the gene either strongly promoted (pro-CML) or strongly opposed (anti-CML) leukemia. To coarse grain the large number of resulting of genes per lineage, we applied weighted gene coexpression network analysis (WGCNA) to define gene modules and module eigengenes that define coordinated transcriptional programs. Each module produced by this process were strongly enriched for either pro- or anti-CML which suggests that they define functional units in leukemia development. We then inferred gene regulatory networks for these modules using Bayesian network inference constrained by prior knowledge from curated interaction and regulatory databases. This produced module-level networks that were unique for each of the B, T, myeloid, and stem cell compartments. For each inferred network, we computed steady states (attractors) and projected the stable transcriptional configurations into the state-space to determine whether the gene derived attractors align with lineage-specific macrostates in the state-space. Preliminary analyses reveal that module networks can reproduce the early, transitional, and late CML macrostates observed from our previous study. Further, we performed in silico perturbations of the networks to predict shifts in attractor occupancy and recapitulate our previous findings that the dominant contributions of B and myeloid compartments to disease progression observed previously. These results support a mechanistic view of leukemia where CML macrostates arise from cell type-specific teams of genes organized into low-dimensional regulatory networks. These network level attractor states could provide a new approach to identify therapeutic targets that are directly related to disease phenotypes and, therefore, new approaches for preventing CML disease evolution. Citation Format: David Eugene Frankhouser, Anupam Dey, Jennifer Rangel Ambriz, Ziang Chen, Denis O'Meally, Yu-Hsuan Fu, Jihyun Irizarry, Tiffany Kanesa Ybarra, Ryan Sathianathen, Jeffrey Trent, Stephen J. Forman, Kathleen M. Sakamoto, Ya-Huei Kuo, Bin Zhang, Adam L. MacLean, Guido Marcucci, Russell Rockne. State-transition model of time-series single-cell RNA-seq identifies gene-level origins of disease microstate stability in chronic myeloid leukemia (CML) [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 6830.

Abstract Waiting for test results is a primary cause of delay in the diagnostic classification and risk stratification of patients with AML. With conventional cytogenetic, FISH, and targeted NGS approaches taking a median turnaround time (TAT) of 7-15 days, there is a significant unmet need to deliver these results faster using a more comprehensive testing platform. We have very recently developed, and demonstrated Analytical Validity (AV), Clinical Validity (CV) and Clinical Utility (CU) in accordance with MolDX L38047, for a rapid Whole Genome Sequencing (WGS) approach for AML (DEX Z-Code Z04C0). Importantly, this test (termed ALTseq) was specifically designed to deliver clinically actionable genomic results that encompass those from cytogenetics, FISH, and targeted NGS in under 48 hours (TAT mean = 33 hours). This accelerated turnaround time was enabled by streamlining laboratory workflows, enhancing bioinformatic pipelines, and expediting variant approval for reporting. The assay captures single nucleotide variants, indels (including FLT3-ITDs and KMT2A-PTDs), and 155 distinct structural variants including KMT2A rearrangements, and genome-wide copy number alterations. Additionally, we have recently developed a method to measure monosomal and complex karyotypes from WGS sequencing data. By deploying WGS we are able to gain insight in all genes in the human exome, with the curated clinical reporting covering all current key AML-related aberrations. These include mutations in NPM1, TP53, RUNX1, IDH1/2, FLT3, MEN1, rearrangements involving KMT2A, MECOM, NUP98, and canonical translocations such as PML::RARA, RUNX1::RUNX1T1, and BCR::ABL1. ALTseq has a limit of detection of 9%, 8%, 10%, and 7% for SNVs, indels, CNAs, and SVs, respectively, with sensitivities of 96%, 96.4%, 95.7%, and 100%, respectively. The positive predictive value for all variant types is ≥99.5%. Since implementation, ALTseq has been used in 66 AML cases, achieving a mean TAT of 33 hours from sample receipt to report delivery. Recent evidence of clinical utility includes the incorporation of Mylotarg in induction therapy based on the identification of CBFB::MYH11 fusion gene, and the inclusion of Revumenib for a KMT2A rearrangement that could not be identified with standard breakaway FISH probes. In summary, we will describe the clinical deployment of a high-throughput, fast-turnaround WGS platform for AML, capable of delivering comprehensive genomic profiling in <48 hours from sample receipt to clinical reporting enabling earlier, more informed treatment decisions. Citation Format: Wayne M. Jepsen, Jonathan J. Keats, Sara A. Byron, Cherie Wesley, Bryce Turner, Christophe Legendre, Tyler Izatt, Tracey White, Amy Stouffer, Lucy Ghoda, Yeneka Campana, Courtney Holden, Jonathan Beteran, Michelle Afkhami, Anthony Stein, Tibor Kovacsovics, Guido Marcucci, Jeffrey Trent. Rapid clinical diagnostic classification and risk stratification in acute myelogenous leukemia (AML) using whole genome sequencing (WGS) [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 2495.

Abstract Extrachromosomal DNA (ecDNA) drives oncogene amplification, transcriptional deregulation, and therapeutic resistance in cancer, but its prevalence and impact in melanoma remain unclear. We analyzed 470 TCGA-SKCM samples with whole-genome (n = 223), whole-exome (n = 247), and RNA sequencing (n = 469). Tumors were classified as BRAF/NRAS/NF1-mutant (BNN, n = 396) or triple wild-type (TWT, n = 74). ecDNA detection was performed using AmpliconArchitect (AA) and Gene-level Circular Amplicon Prediction (GCAP). Multi-omic analyses assessed transcriptional, immunologic, and clinical correlates. ecDNA was detected in 133 tumors (28%; 47 by AA, 133 by GCAP, 39 overlapping), with higher prevalence in TWT tumors (51%) than BNN tumors (26%; χ2 = 17.88, p = 2.4 × 10−5). Frequently amplified ecDNA genes included MDM2, CDK4, CCND1, BIRC2/3, PAK1, GAB2, and RSF1, implicating proliferation, apoptosis evasion, and chromatin regulation. In an effort to assess putative functional impact, we performed various transcriptomic analyses which identified upregulation of amplified oncogenes (GAB2, MDM2, RSF1, CCND1, PAK1). In addition, gene set enrichment analyses with MutSigDB (against Hallmark genes) shows strong enrichment of pathways associated with enhanced cell proliferation, such as MYC and E2F targets in ecDNA+ tumors. In addition, cell deconvolution methods reveal immune cell composition varies across subtypes, highlighting potential unique immunomodulatory effects. Survival analysis indicated worse overall survival in ecDNA+ tumors (p = 0.016), driven by TWT cases (p = 0.0048). ecDNA is prevalent in melanoma, particularly in TWT tumors, where it promotes oncogene amplification, immunomodulatory reprogramming, and poor prognosis. These results highlight ecDNA as a potential biomarker for stratification and targetable vulnerability in TWT melanoma. Citation Format: Sharadha Sakthikumar, Bryce Turner, Jeffrey Trent, ALEKSANDAR SEKULIC. Integrative multi-omic analysis reveals subtype-specific impact of extrachromosomal DNA in melanoma [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 1994.

Abstract Background: Multi-cancer early detection (MCED) from plasma liquid biopsy has advanced rapidly, demonstrating that diverse cfDNA features can reveal early tumor signals. Most current assays rely on whole-genome or methylation sequencing, which are costly and require deep coverage. EarlySeek is a novel, highly multiplexed cfDNA amplicon-based alternative that targets ∼800,000 SINE-enriched loci using long and short amplicons, generating a bimodal insert-size distribution and requiring as little as 0.25 ng of DNA input. This design captures complementary signals such as fragmentation, aneuploidy, genomic abundance shifts, and sequence motif patterns. We evaluated a multi-signal framework integrating these features using artificial intelligence and machine learning. Methods: Three datasets were analyzed: a training set (237 cancers, 463 normals), a calibration set (72 cancers, 140 normals), and an independent test set with tissues present in training (78 cancers, 192 normals). EarlySeek output yields six biological scores: fragment length, two aneuploidy scores, two coverage/abundance scores, and a 6-mer motif score. Scores were generated using representation and deep learning architectures, including autoencoders that compress amplicon-level and binned genomic information into informative latent features, combined with machine learning classifiers such as support vector machines and gradient-boosted trees. Each score was calibrated via quantile-to-quantile regression, and a predefined multi-signal rule was applied to generate final scores for the test cohort. Results: Across 270 independent test samples, the combined six-score framework achieved 51% sensitivity at 99% specificity (95% CI: 40%-62%). Sensitivity by stage showed meaningful early detection: 45% for stage I (CI 26%-66%), 58% for stage II ( CI 39%-74%), 57% for stage III ( CI 37%-67%), and 100% for stage IV (2/2; CI 34%-100%). Performance varied by cancer type, with strongest detection in colorectal (61%), gastric (71%), liver (80%), ovarian (75%), and pancreatic cancer (64%). Lower sensitivities in breast, and prostate cancers reflected known low cfDNA shedding. Conclusions: EarlySeek’s bimodal amplicon design enables extraction of diverse cfDNA signals from a single sequencing assay. Integrating these orthogonal features through artificial intelligence and machine learning supports high-specificity MCED detection and yields meaningful early-stage performance. This approach offers a scalable, cost-effective MCED test, a desirable feature for population-level cancer screening. Citation Format: Kamel Lahouel, Kameron Bates, Victoria Zismann, Candice Wike, Kunjur Manasa Upadhyaya, Matteo Munini, Mete Mulazimoglu, Gracyn Benck, Kianna Martos Rupp, Payton Smith, Chaney Jambor, Sophie Pénisson, Stephanie Pond, Jeffrey Trent, Cristian Tomasetti. Liquid biopsy MCED enabled by a novel 800K-locus bimodal amplicon sequencing technology [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 7624.

Small cell carcinoma of the ovary, hypercalcemic type (SCCOHT), is a rare, deadly form of ovarian cancer that uniformly harbors mutations in SMARCA4, a member of the SWI/SNF chromatin remodeling complex. SWI/SNF impacts RNA splicing, and dysregulation of splicing can generate immunogenic tumor antigens. In this study, we explored the relationship between SMARCA4 loss and RNA splicing dysregulation. SCCOHT primary tumors harbored tumor-associated outlier splicing events compared with normal tissues. Many of the tumor events were retained introns encoding novel peptides predicted to bind to MHC-I complexes. Immune cells were observed in primary SCCOHT tumors, suggesting a potentially immune-reactive tumor microenvironment. Mutations in several switch/sucrose nonfermenting (SWI/SNF) subunits were associated with higher rates of outlier retained introns across tumor types in The Cancer Genome Atlas data. Interestingly, RNA sequencing of isogenic SCCOHT cell lines demonstrated a role for SMARCA4 in intron retention (IR). Distinct protein-protein interactions between splicing factors identified in SCCOHT cell lines supported a role for SMARCA4 in splicing regulation. Furthermore, SWI/SNF localized to genes, which were differentially spliced. Mass spectrometry analyses confirmed expression of some of these novel peptides, and a subset of these are predicted to bind to MHC-I complexes. A pool of these novel peptides derived from retained introns in SCCOHT triggered proliferation and expression of TNFα and INFγ in primary human T cells. Together, these data suggest that SMARCA4 loss in SCCOHT leads to IR. Furthermore, T-cell activation by novel peptides encoded by these tumor-specific splicing events suggests IR could be a source of tumor-associated antigens in SCCOHT. SIGNIFICANCE:SCCOHT, a rare ovarian cancer, features splicing dysregulation due to SMARCA4 loss that generates immunostimulatory peptides linked to potential immune responses and therapeutic avenues, challenging traditional views of the role of SMARCA4.

Abstract CML progression is defined by increased disease burden and potential for transformation to blastic phase (BP), yet the mechanisms and ability to predict this evolution remain elusive. We applied elements of state-transition theory to model transcriptional dynamics underlying leukemic progression, where individual cells' gene expression profiles represent microstates, and aggregated population-level gene expression profiles define macrostates that govern phenotypic transitions from health to disease initiation to overt leukemia. Previously, we demonstrated through bulk transcriptomic analyses that CML progression and therapeutic responses can be accurately predicted by modeling a leukemic potential landscape derived from transcriptional state-space geometry. To extend this framework to single-cell (sc) resolution and capture more granular information at the microstate level, we collected weekly peripheral blood samples from inducible chronic phase (CP) and BP CML mouse models to generate time-series scRNA-seq data tracking the transition from health to overt CP or BP. After performing quality control and cell labeling, we used the first (healthy) time point before the induction of BCR::ABL (T0) and the final leukemia time point (Tf), in addition to other intermediate time points, to assess changes in cell populations and gene expression. All analyses were first performed in CP mice, validated in BC mice, and corroborated using two independent human CML scRNA-seq datasets. By comparing sequential time points, we observed that progression to CP CML was associated with a significant decrease in B cells (p<0.01) and increases in both myeloid (p<0.01) and stem cell (p<0.001) populations. Differential gene expression (DEG) analysis revealed marked transcriptional changes across all major lineages (DEGs: B cells = 781; T cells = 2,149; Myeloid = 1,999; Stem cells = 194). Despite these changes, no single-cell transcriptional state at any time point was uniquely associated with health or disease, as transcriptional profiles of leukemic cells largely overlapped with those from healthy cells. However, when scRNA-seq data were computationally aggregated into pseudobulk (PsB) samples, mimicking bulk RNA-seq, a distinct CML state-space emerged, revealing a clear disease trajectory defined by three stable macrostates representing early, transitional, and late leukemia. Cell type–specific PsB analyses uncovered independent state-transition dynamics in B cell, T cell, myeloid, and stem cell compartments, highlighting previously unrecognized complexity in disease evolution involving multiple lineages. We further showed that the transition from health to leukemia at the PsB level could be reconstructed as a linear combination of transitions within individual cell subpopulations. To this end, we quantified each lineage's contribution to disease progression by performing computational simulations that subtracted the influence of each cell type in turn. Interestingly, B and myeloid populations contributed most and comparably to the global disease trajectory—despite B cells decreasing and myeloid cells expanding over time. This counterintuitive finding suggests that leukemic information is encoded not simply by abundance but by dynamic transcriptional shifts across compartments. These findings were validated in a BP mouse model and in two human CML datasets. In summary, we introduce a conceptual framework that distinguishes between microstates (individual cells) and macrostates (population-level transcriptional ensembles), showing that disease progression is not encoded at the single-cell level but instead emerges only when cells are aggregated. While this may appear intuitive, we were surprised to find that higher resolution sc data did not provide clearer insight than pseudobulk data. We speculate that this may reflect greater Shannon information entropy at the single-cell level, which introduces variability that obscures coherent state transitions. To our knowledge, this framework is novel and may inform future approaches to interpreting scRNA-seq data in leukemia and other diseases.

Small cell carcinoma of the ovary, hypercalcemic type (SCCOHT) is a rare, deadly form of ovarian cancer that uniformly harbors mutations in SMARCA4, a member of the SWI/SNF chromatin remodeling complex. SWI/SNF impacts RNA splicing, and dysregulation of splicing can generate immunogenic tumor antigens. Here, we explored the relationship between SMARCA4 loss and RNA splicing dysregulation. SCCOHT primary tumors harbored tumor-associated outlier splicing events compared to normal tissues. Many of the tumor events were retained introns encoding novel peptides predicted to bind to MHC-I complexes. Immune cells were observed in primary SCCOHT tumors, suggesting a potentially immune reactive tumor microenvironment. Mutations in several SWI/SNF subunits were associated with higher rates of outlier retained introns across tumor types in TCGA data. Interestingly, RNA sequencing of isogenic SCCOHT cell lines demonstrated a role for SMARCA4 in intron retention. Distinct protein-protein interactions between splicing factors identified in SCCOHT cell lines supported a role for SMARCA4 in splicing regulation. Further, SWI/SNF localized to genes which were differentially spliced. Mass spectrometry analyses confirmed expression of some of these novel peptides and a subset of these are predicted to bind to MHC-I complexes. A pool of these novel peptides derived from retained introns in SCCOHT triggered proliferation and expression of TNFa and INFb in primary human T cells. Together, these data suggest that SMARCA4 loss in SCCOHT leads to intron retention. Furthermore, T cell activation by novel peptides encoded by these tumor-specific splicing events suggests intron retention could be a source of tumor-associated antigens in SCCOHT. Elizabeth Raupach, Apurva Hegde, Krystine Garcia-Masnfield, Marice Alcantara, David Rose, Rebecca Halperin, Krystal Orlando, Jessica Lang, Ritin Sharma, Victoria David-Dirgo, Salvatore Facista, Rayvon Moore, Rochelle Kofman, Zoe Jensen, Victoria Zismann, Anthony Karnezis, Yemin Wang, Lynda Bennett, Timothy Whitsett, Marcin Kortylewski, William Hendricks, David Huntsman, Lorna Rodriguez-Rodriguez, Bernard Weissman, Jeffrey Trent, Patrick Pirrotte. Loss of SMARCA4 leads to intron retention and generation of tumor-associated antigens in small cell carcinoma of the ovary, hypercalcemic type [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Advances in Ovarian Cancer Research; 2025 Sep 19-21; Denver, CO. Philadelphia (PA): AACR; Cancer Res 2025;85(18_Suppl):Abstract nr B065.

543 Background: We previously reported that combining CBM588 ( Clostridium butyricum MIYAIRI588), a live bacterial product, with cabozantinib (cabo) and nivolumab (nivo) enhanced clinical benefit in treatment-naïve patients with mRCC (Ebrahimi et al ; Nature Medicine 2024). The current study provides updated clinical data to further evaluate the potential benefits of CBM588 in combination with cabo/nivo. Methods: This open-label, randomized trial enrolled patients aged ≥18 years old with a Karnofsky performance status ≥70% and histologically verified (clear-cell, papillary or sarcomatoid component) advanced or mRCC with no prior systemic therapy for metastatic disease. Patients were randomized in a 1:2 ratio to receive either cabo/nivo (40mg PO QD and 480mg IV monthly, respectively) alone or with CBM588 (80mg PO BID). This analysis provides updated secondary clinical endpoints with extended follow-up, including overall response rates (ORR), progression-free survival (PFS), and toxicity. Clinical benefit was defined as complete response, partial response, or stable disease, per RECIST 1.1. The association between treatment arm and ORR was evaluated using Fisher’s exact test, and PFS was estimated using the Kaplan-Meier method. Results: A total of 30 patients (20:10 M:F) were recruited, with a median age of 65 years (range, 36-84). Five patients (17%) had sarcomatoid features, and two (7%) had predominant papillary histology. As of June 1, 2024, the median follow-up was 25.8 months (interquartile range, 19.2-28.1) in the overall cohort. The ORR was significantly higher in the CBM588-containing arm compared to cabo/nivo alone arm (79% versus 20%, P =0.004). In the CBM588 arm, 17 (89%) patients, and in the control arm, 8 (80%) patients had a reduction in target lesion size, with median decreases of 51% (range, 17-94%) and 22% (range, 13-100%), respectively. Clinical benefit for at least 6 months was achieved in 80% of patients treated in experimental arm and 60% patients in the control arm. The median PFS was not reached in patients receiving CBM588, compared to 13.4 months in the control arm. The median OS was not reached in either of the arms at the time of data cutoff. Grade 3 or higher treatment-related adverse events (TRAEs) were observed in 45% of the CBM588 arm compared to 40% in the control arm. The most common TRAEs in the overall cohort were transaminitis (10%), hypertension (7%), and diarrhea (7%), with no significant differences between treatment arms. No new safety signals were detected. Conclusions: The addition of CBM588 to cabo/nivo continues to show promising efficacy in mRCC, with an improved PFS and ORR. The safety profile remains consistent with previous findings, supporting further exploration in larger trials. Further translational efforts are underway to characterize the mechanism through which CBM588 augments clinical activity. Clinical trial information: NCT05122546 .

4550 Background: In two randomized phase I trials, Clostridium butyricum MIYAIRI588 (CBM588), a live biotherapeutic, demonstrated preliminary activity in modulating the gut microbiome, enhancing systemic immune responses, and improving clinical outcomes in patients receiving first-line nivolumab/ipilimumab and nivolumab/cabozantinib for mRCC (Dizman et al. and Ebrahimi et al. Nature Medicine). Herein, we present the long-term follow-up data for nivolumab/ipilimumab with or without CBM588. Methods: Newly diagnosed patients with mRCC, clear cell and/or sarcomatoid histology, and International mRCC Database Consortium intermediate/high risk were randomized to receive nivolumab/ipilimumab with or without CBM588 in a 2:1 ratio. Response outcomes were assessed using RECIST 1.1. Clinical outcomes were secondary endpoints. Objective response rate (ORR; complete response [CR] or partial response [PR]), disease control rate (DCR; CR, PR, or stable disease [SD] > 6 months), progression-free survival (PFS), and overall survival (OS) outcomes were compared across arms. Results: Twenty-nine patients were included in the final analysis: 19 in the nivolumab/ipilimumab with CBM588 arm and 10 in the nivolumab/ipilimumab arm. The median age was 66.2 years, 72% were male, 83% had IMDC intermediate risk and 93% had clear cell histology. Baseline characteristics were similar across arms. ORR and DCR were 58% and 79% in nivolumab/ipilimumab with CBM588 arm versus 20% and 20% in nivolumab/ipilimumab arm, respectively (p = 0.06 and p = 0.004). At a median follow-up of 60.0 (95% CI 51.9-68.1) months, the median PFS was 38.2 (95% CI 23.6-52.8) months in the nivolumab/ipilimumab and CBM588 arm versus 19.3 (95% CI 0-41.9) months in the nivolumab/ipilimumab arm (Hazard ratio [HR] 0.24, 95% CI 0.09-0.61 p = 0.003). At the time of data cutoff, 9 (47.4%) and two (20%) patients were alive in the nivolumab/ipilimumab with CBM588 and nivolumab/ipilimumab arms, respectively. The median OS with nivolumab/ipilimumab with CBM588 was 55.0 (95% CI 10.5-75.5) months versus 39.0 (95% CI 23.7-54.3) months with nivolumab/ipilimumab (HR 0.438 [95% CI 0.17-1.1] p = 0.09). Conclusions: Although limited by the sample size, the combination of nivolumab/ipilimumab with CBM588 demonstrated superior clinical activity over nivolumab/ipilimumab in our cohort. Additionally, ORR, PFS and OS with nivo/ipi/CBM588 exceeded those observed with nivolumab and ipilimumab in historical datasets (Motzer et al. NEJM). Larger efforts investigating the impact of CBM588 on clinical outcomes are underway. Clinical trial information: NCT03829111 .

402 Background: Nearly one third of men with prostate cancer experience relapse after radical prostatectomy, often despite achieving undetectable PSA nadir. An assay to detect minimal residual disease would be valuable to identify patients who could benefit from adjuvant therapy. Methods: Eligible patients had biopsy proven prostate cancer and planned to undergo prostatectomy for definitive therapy. Blood was drawn at pre-specified timepoints (TP) including prior to prostatectomy (TP1), on the day after surgery or at the 1-week post-op catheter removal visit (TP2), and at 1-month post-op with no interval treatment (TP3). We performed tumor-guided plasma DNA analysis using a novel method based on single-stranded adapter ligation, incorporation of unique molecular identifiers (UMIs), targeted PCR, normalization of amplicons using hybrid-capture, and targeted sequencing. Using tumor and germline exome sequencing, we designed and validated multiplexed assays to target somatic founder mutations. We prepared targeted sequencing libraries using a median of 6.7 ng input plasma DNA. Results: 11 patients were enrolled after IRB approval. 1 (9.09%) had grade group 2, 1 (9.09%) grade group 3, 5 (45.45%) grade group 4, 4 (36.36%) grade group 5 disease. Median PSA pre-op was 9.3 ng/mL (range: 1.7 – 21.47). Tumor DNA sequencing was successful and targeted assays were generated for all patients (11/11). Median number of targets tested was 20 (range 4 to 88). We analyzed 31 plasma DNA samples obtained from 11 patients. Median on-target rate in plasma DNA was 90.5% (range 76.0% to 96.3%). Prior to surgery (TP1), 5/11 (45%) patients had detectable ctDNA at a median tumor fraction of 0.077% (range: 0.006% to 3.4%). At post-op TP (TP2), the ctDNA was detectable in 7/11 patients (64%) at a median tumor fraction of 0.024% (range: 0.004% to 3.1%). Detectable ctDNA was observed at 1 month (TP3) in 5/8 patients (62.5%) at a median tumor fraction of 0.022% (range 0.010% to 0.037%). For an exploratory analysis, we examined whether persistent ctDNA detection at the 1 month TP3 was associated with later relapse. Of the 5 patients with PSA relapse, 3 (60%) of patients had detectable residual ctDNA at TP3 with median follow up time of 61.75 months. Conclusions: Localized prostate cancer has been noted to have low tumor shed, limiting meaningful ctDNA detection rates with prior assays (Hennigan 2019). Our results show promising feasibility of a novel tumor-informed ctDNA assay in this setting. In addition, there was preliminary evidence of residual ctDNA detection predicting relapse.

Small cell carcinoma of the ovary-hypercalcemic type (SCCOHT) is a rare ovarian cancer affecting young females and is driven by the loss of both SWI/SNF ATPases SMARCA4 and SMARCA2. As loss of SWI/SNF alters enhancers, we hypothesized that super-enhancers, which regulate oncogene expression in cancer, are disparately impacted by SWI/SNF loss. We discovered differences between SWI/SNF occupancy at enhancers vs. super-enhancers. SCCOHT super-enhancer target genes were enriched in developmental processes, most notably nervous system development. This may further support neuronal cell-of-origin previously proposed. We found high sensitivity of SCCOHT cell lines to triptolide. Triptolide inhibits expression of many super-enhancer-associated genes, including oncogenes. SALL4 expression is decreased by triptolide and is highly expressed in SCCOHT tumors. In patient-derived xenograft models, triptolide and prodrug minnelide effectively inhibit tumor growth. These results reveal unique features of super-enhancers in SCCOHT, which may be one mechanism through which triptolide has high activity in these tumors.

Analysis of circulating tumor DNA (ctDNA) using tumor-guided bespoke sequencing can achieve high accuracy for minimal residual disease detection. However, bespoke laboratory workflows are often complex, difficult to scale, and costly. Multiplexed PCR methods have lower costs and preserve input DNA but instill higher background noise, variable on-target rates and unpredictable relative representation of amplicons. Locus- and mutation-specific hybrid capture can overcome these challenges but with higher costs of patient-specific custom baits and greater loss of input DNA. Here, we describe a workflow that combines targeted amplification with fixed content exome-wide hybrid capture to overcome these challenges. We performed tumor-guided plasma DNA analysis using a novel method based on enrichment and normalization of amplicons and targeted sequencing. Using tumor and germline exome sequencing, we identified somatic founder mutations, and designed primers for multiplexed amplification. Targeted sequencing libraries were prepared using a combination of single-stranded adapter ligation, targeted PCR and hybrid capture. Sequencing reads were grouped in read families based on fragment size and unique molecular identifiers, and analyzed to improve specificity for detection of known mutations. We evaluated assay performance using reference samples (Seracare Life Sciences) as well as plasma samples from patients with cancer. We analyzed 8 replicates of reference samples at 10 ng input each for variant allele fractions (VAFs) of 0.5%, 0.25% and 0.125%, using an 18 mutation panel, and detected mutated DNA in 8/8 samples for all VAFs. In addition, we prepared a dilution at 0.01% VAF and analyzed 16 replicates of this sample at 40 ng input each, and detected mutated DNA in 15/16 samples. Using the same analytical thresholds, 1/8 wild-type reference samples were called positive. A comparison of observed and expected tumor fractions across all samples showed a correlation coefficient of 0.97 (p<0.001, n=48). In addition, we developed 44 patient-specific panels targeting a median of 30 mutations per patient (range 5 to 125 mutations) for patients with pancreatic cancer, breast cancer and glioblastoma. Using a median input of 6.7 ng (range 1.1 to 24 ng), we analyzed 112 plasma DNA samples across 44 panels and observed a median on-target rate of 89% using the first round of panel design. ctDNA was detected in 62 plasma samples (55%) at a median tumor fraction of 0.013% (range 0.009% to 5.8%). Our results demonstrate analytical performance and assay stability across 44 patients using bespoke sequencing of enriched amplicons for circulating tumor DNA (enACT-DNA) analysis. On-going studies are evaluating the clinical relevance of perioperative ctDNA detection using enACT-DNA in patients with cancer. Tania Contente-Cuomo, Devin Dinwiddie, Mojca Stampar, Adriana Marshall, Michael E. Berens, Stephanie J. Pond, Jeffrey M. Trent, Muhammed Murtaza. Bespoke sequencing of enriched and normalized amplicons increases on-target rate and reduces background noise for circulating tumor DNA detection and quantification [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 4562.

AML accounts for one-third of adult leukemias. With an incidence of approximately 22, 000 new cases/year, AML is projected to cause over 11,000 deaths in the United States in 2025. Hence, new and more effective diagnostic, prognostic, and therapeutic approaches are highly needed. Genomic profiling is crucial to provide a proper prognosis and predict treatment response to emerging new therapeutics for both younger adult (<60 y/o) or older (>60 y/o) AML patients. This information is key to determine the best course of treatment and to make earlier decisions of whether to proceed to allogeneic hematopoietic cell transplants. However, the turnaround time for traditional “gold-standard” diagnostics — including FISH, cytogenetics, and targeted gene panels — can often exceed two weeks, depending on laboratory capacity and healthcare system logistics. Because many patients with AML require immediate initiation of therapy, clinicians are frequently forced to make treatment decisions without the benefit of genomic insights that could optimize therapy selection or trial eligibility — particularly in patients with high-risk, refractory, or relapsed disease. To address this unmet need, we developed ALTseqTM, a rapid whole genome sequencing (rWGS) assay for AML, at the TGen Clinical Laboratory (CAP: 8713783, CLIA: 03D2215886). ALTseqTM was specifically designed to deliver clinically actionable genomic results in under 48 hours, significantly faster than conventional testing. This accelerated turnaround time was enabled by streamlining laboratory workflows, enhancing bioinformatic pipelines, and expediting variant approval for reporting. The assay captures single nucleotide variants (SNVs), insertions and deletions (indels) — including FLT3-ITDs and KMT2A-PTDs —, 155 distinct structural variants (SVs including translocations or inversions) along with KMT2A rearrangements, and genome-wide copy number alterations, all in a single test. Specimens received must be whole blood or bone marrow in an EDTA tube and contain ≥20% blasts. DNA is extracted using the Qiagen DNA Blood Mini Kit (Qiagen, Hilden, Germany), prepared using the Watchmaker Genomics DNA Library Prep Kit (Watchmaker Genomics, Denver, CO, USA), and sequenced on the NovaSeq X Plus (Illumina, San Diego, CA, USA). The analytical pipeline utilizes both commercially available and proprietary algorithms with resulting VCFs compared to a knowledge base built in-house to create a json file used by the LIMS to produce the final clinical report. The curated report covers the genomic status of 41 genes with validated detection of key AML-related aberrations in both peripheral blood and bone marrow specimens. These include mutations in NPM1, TP53, RUNX1, IDH1/2, FLT3, MEN1, rearrangements involving KMT2A, MECOM, NUP98, and canonical translocations such as PML::RARA, RUNX1::RUNX1T1, and BCR::ABL1, among others. ALTseqTM has a limit of detection of 9%, 8%, 10%, and 7% for SNVs, indels, CNAs, and SVs, respectively, with sensitivities of 96%, 96.4%, 95.7%, and 100%, respectively. The positive predictive value for all variant types is ≥99.5%. Since implementation, ALTseqTM has been used in 35 AML cases, achieving an average turnaround time (TAT) of 35.2 hours, with a range of 29.4 to 46.3 hours from sample receipt to report delivery. Physician satisfaction with the assay's speed and clinical utility has been consistently high at City of Hope where the initial launch has occurred. In summary, we describe the successful deployment of a high-throughput, fast-turnaround rWGS platform for AML, capable of delivering comprehensive genomic profiling of 41 genes in <48 hours from sample receipt to clinical reporting. This approach has enabled earlier, more informed treatment decisions, and we are actively evaluating its impact on hospital length of stay and clinical trial enrollment, particularly where molecular eligibility criteria are required. Building on this success, we are now validating a similar rWGS approach for patients with multiple myeloma and plan to expand ALTseqTM to include acute lymphoblastic leukemia (ALL).

PURPOSE:NCI selected a network of Clinical Laboratory Improvement Amendments-certified laboratories performing routine next-generation sequencing (NGS) tumor testing to identify patients for the NCI Molecular Analysis for Therapy Choice (NCI-MATCH) trial. This large network provided a unique opportunity to compare variant detection and reporting between a wide range of testing platforms. EXPERIMENTAL DESIGN:Twenty-eight NGS assays from 26 laboratories within the NCI-MATCH Network, including the NCI-MATCH central laboratory (CL) and 11 commercial and 14 academic designated laboratories (DL), were used for this study. DNA from eight cell lines and two clinical samples were sequenced. Pairwise comparisons in variant detection and reporting between each DL and CL were performed for single-nucleotide variant, insertion and deletion, and copy-number variant classes. RESULTS:We observed high concordance in variant detection between CL and DL for single-nucleotide variants and insertions and deletions [average positive agreement (APA) > 95.4% for all pairwise comparisons] but lower concordance for variant reporting after analysis pipeline filtering. We observed much higher agreement between CL and assays using amplification as the target enrichment method (84.2% < APA ≤ 95.7%, average APA = 88.7%) than other assays using hybridization capture (69.7% < APA ≤ 93.8%, average APA = 77.4%) due to blacklisting of actionable variants in low complexity regions. For copy-number variant reporting, we observed high agreement (APA > 82%) except between CL and two assays (APA = 76.9% and 71.4%) due to differences in estimation of copy numbers. Notably, for all variants, differences in variant interpretation also contributed to reporting discrepancies. CONCLUSIONS:This study indicates that different NGS tumor profiling tests currently in widespread clinical use achieve high concordance between assays in variant detection. For variant reporting, observed discrepancies are mainly introduced during the bioinformatic analysis.

PURPOSE:We describe strategies implemented across research centers of the Participant Engagement and Cancer Genome Sequencing (PE-CGS) Network to optimize engagement of participants and communities in cancer genomics research. We also present consensus definitions of engagement and engagement optimization, informed by our shared experiences in the Network. METHODS:Key informant interviews and a document review identified engagement and optimization strategies across PE-CGS research centers. Findings were synthesized using qualitative content analysis. Consensus on definitions of engagement and optimization were developed through iterative review by PE-CGS members. RESULTS:PE-CGS research centers adopted tailored strategies based on community needs and scientific gaps. Engagement strategies included community-based efforts (eg, advisory boards and newsletters) and participant-focused approaches (eg, enhanced informed consent and decision support tools). Optimization strategies leveraged scientific methods (eg, randomized controlled trials and surveys) to evaluate engagement. Engagement was described as the sustained and meaningful interactions between researchers, participants, and communities. Optimization was described as the application of scientific methods to refine and improve engagement and research processes and outcomes. CONCLUSION:Engagement and optimization strategies have informed research planning, conduct, and dissemination across PE-CGS. These approaches and definitions provide a foundation for developing evidence-based practices to strengthen participant and community involvement in cancer genomics research.

We reported that an acquired miR-142 deficit transforms chronic phase (CP) chronic myeloid leukemia (CML) leukemic stem cells (LSCs) into blast crisis (BC) LSCs. Given the role of miR-142 in the development and activity of the immune system, we postulated that this deficit also promotes LSC immune escape. Herein, we report on IL-6-driven miR-142 deficit occurring in T cells during BC transformation. In CML murine models, miR-142 deficit impairs thymic differentiation of lymphoid-primed multipotent progenitors (LMPP) into T cells and prevents T cells' metabolic reprogramming, thereby leading to loss of T cells and leukemia immune escape. Correcting miR-142 deficit with a miR-142 mimic compound (M-miR-142), alone or in combination with immune checkpoint antibodies, restores T cell number and immune activity, leading to LSC elimination and prolonged survival of BC CML murine and patient-derived xenograft models. These observations may open new therapeutic opportunities for BC CML and other myeloid malignancies.

Background Children with relapsed central nervous system (CNS tumors), neuroblastoma, sarcomas, and other rare solid tumors face poor outcomes. This prospective clinical trial examined the feasibility of combining genomic and transcriptomic profiling of tumor samples with a molecular tumor board (MTB) approach to make real‑time treatment decisions for children with relapsed/refractory solid tumors. Methods Subjects were divided into three strata: stratum 1—relapsed/refractory neuroblastoma; stratum 2—relapsed/refractory CNS tumors; and stratum 3—relapsed/refractory rare solid tumors. Tumor samples were sent for tumor/normal whole-exome (WES) and tumor whole-transcriptome (WTS) sequencing, and the genomic data were used in a multi-institutional MTB to make real‑time treatment decisions. The MTB recommended plan allowed for a combination of up to 4 agents. Feasibility was measured by time to completion of genomic sequencing, MTB review and initiation of treatment. Response was assessed after every two cycles using Response Evaluation Criteria in Solid Tumors (RECIST). Patient clinical benefit was calculated by the sum of the CR, PR, SD, and NED subjects divided by the sum of complete response (CR), partial response (PR), stable disease (SD), no evidence of disease (NED), and progressive disease (PD) subjects. Grade 3 and higher related and unexpected adverse events (AEs) were tabulated for safety evaluation. Results A total of 186 eligible patients were enrolled with 144 evaluable for safety and 124 evaluable for response. The average number of days from biopsy to initiation of the MTB-recommended combination therapy was 38 days. Patient benefit was exhibited in 65% of all subjects, 67% of neuroblastoma subjects, 73% of CNS tumor subjects, and 60% of rare tumor subjects. There was little associated toxicity above that expected for the MGT drugs used during this trial, suggestive of the safety of utilizing this method of selecting combination targeted therapy. Conclusions This trial demonstrated the feasibility, safety, and efficacy of a comprehensive sequencing model to guide personalized therapy for patients with any relapsed/refractory solid malignancy. Personalized therapy was well tolerated, and the clinical benefit rate of 65% in these heavily pretreated populations suggests that this treatment strategy could be an effective option for relapsed and refractory pediatric cancers. Trial registration ClinicalTrials.gov, NCT02162732. Prospectively registered on June 11, 2014.