Citations

With the rapid development of artificial intelligence (AI), various deep learning (DL) methods have been introduced into radiation oncology. Among them, the generation of synthetic Computed Tomography (sCT) images has attracted increasing attention, as it supports different clinical scenarios, from image-guided adaptive radiotherapy (IGART) to the simulation-free workflow. This review provides a comprehensive overview of recent studies on DL-based sCT synthesis in radiotherapy from multiple imaging modalities, including Cone-Beam CT (CBCT), Magnetic Resonance Imaging (MRI), and diagnostic CT, and discusses their clinical applications in CBCT-based online adaptive radiotherapy, MRI-guided radiotherapy, and simulation-free workflows. We also examine the architectures of representative DL models such as convolutional neural networks (CNNs) and generative adversarial networks (GANs) and summarize emerging training strategies. Finally, we discuss current challenges of clinical translation of DL algorithms into clinical practice and suggest potential directions for future research. Overall, this paper highlights the potential of AI-driven sCT generation to advance treatment planning by reducing imaging burden, improving dose accuracy, and accelerating workflow efficiency, thus ultimately improving the treatment outcome of patient care.

Context.— Clinical laboratories are increasingly implementing pharmacogenomic (PGx) testing. Although PGx is similar to genetic testing for other indications, there are unique aspects that laboratories should consider. Objective.— To aid clinical laboratories that are implementing clinical PGx testing by describing characteristics of PGx test design and validation, as well as approaches to reporting. Resources that are useful for clinical laboratories performing PGx testing will be highlighted. Design.— The College of American Pathologists formed a workgroup composed of laboratorians with expertise in clinical PGx testing. The workgroup included representatives from the Association for Molecular Pathology and the American College of Medical Genetics and Genomics. The workgroup reviewed pertinent literature, as well as experience from proficiency testing and from members’ laboratories. Results.— The workgroup recommends that laboratories implementing PGx consider the following concepts: testing platform, test design (ie, selection of pharmacogenes and variants/alleles), use of reference materials during test development and as controls during clinical runs, star allele and standard nomenclature systems, translations from genotype to predicted phenotype, and considerations for result reporting including making medication recommendations. The workgroup provides considerations when using report vendors, emphasizing the clinical laboratory’s role and responsibility when implementing such reporting tools from vendors. Conclusions.— Clinical laboratories should be familiar with the fundamentals of PGx, ensure that PGx testing meets the applicable regulatory requirements for all aspects of the clinical laboratory testing process, and follow recommendations for standardization of nomenclature and reporting.

Abstract Description Traditional methods for identifying neoantigen-specific TCRs rely on labor-intensive in vitro co-cultures, which can alter T cell characteristics. We aim to identify endogenous neoantigen-specific TCRs in their natural state using antigen-mapping single-cell RNA and TCR sequencing. We profiled circulating CD8+ T cells from patients with KRAS-mutant metastatic colorectal adenocarcinoma (CRA) to identify TCRs that target HLA-A restricted (*02:01 or *11:01) mutant KRAS epitopes (G12D or G12V) using Barcode Enabled Antigen Mapping (BEAM; 10x Genomics) and performed cluster analysis to decipher T cell function. CD8+ T cells were isolated from peripheral blood and stained with peptide-MHC complexes for KRAS G12D/V epitopes (HLA-A02:01 and HLA-A11:01). Sequencing data were processed with Cell Ranger and Seurat. We analyzed 12 specimens from 6 patients who were receiving fluorouracil-based chemotherapy for metastatic CRA and detected neoantigen-specific TCR clonotypes in 11 specimens, suggesting their persistent presence in peripheral blood. TCR clonotypes displayed varying functional states, with some showing cytotoxic effector markers (GZMB, NKG7) and others showing markers of Naïve T cells (LEF1, TCF7, SELL, CCR7). These findings show that neoantigen-specific TCRs have diverse functional states and evolve over time. This approach is a promising tool for identifying endogenous neoantigen-specific TCRs and provides insights into TCR structure and function for adoptive therapy. Funding Sources This project is supported by the Eric & Wendy Schmidt Fund for AI Research & Innovation at Mayo Clinic (MZ and HD), and Mayo Clinic Comprehensive Cancer Center Distinguished Senior Scientist Award (HD). Topic Categories Tumor Immunology: Checkpoints, Prevention, and Treatment (TIPT)

Abstract Description A durable T cell immunity against cancer is dependent on a continually refilling of effector CD8+ T cells. Thymic output has been correlated with favorable prognosis in cancer patients of advanced ages, suggesting thymus is an important source for refilling T cells that are capable of controlling cancer progression in the periphery. However, it remains unclear how PD-1 signals regulate the cytotoxic potential of CD8+ T cells in thymus before they egress to the periphery to establish durable anti-tumor immunity. Here we report PD-1 signaling prelimits thymic single positive (SP) CD8+ T cells to acquire cytotoxic potential by downregulating Gzmb and Nkg7, which are required for CD8+ T cell-mediated cytotoxicity. Unexpectedly, we found the absence of Pdcd1 also led to upregulation of Tox and Tim3 in thymic SP CD8+ T cells, thereby pre-programming these cells for exhaustion within the tumor. Thus, the absence of PD-1 signals in CD8+ T cells only partially delayed tumor growth in peripheral tissues. To improve therapeutic effects of PD-1 inhibition, we observed that tumor growth in CD8-specific Pdcd1 conditional knockout mice was further suppressed by treatment with a PD-L1 antibody that blocks PD-L1 and CD80 interaction. Our results imply that PD-1 inhibition promotes the expansion of effector cytotoxic CD8+ T cells but also retains T cell exhaustion potential, which can be mitigated by PD-L1 blockade that allows more CD80 to provide costimulatory signals to T cells via CD28. Funding Sources NCI R01CA256927 NCI R01CA261932 NCI P30CA15083 Topic Categories Tumor Immunology: Cellular Responses and Tumor Microevironment (TIME)

Abstract Description A subset of follicular lymphoma patients has an early relapse with a poor outcome. We showed that this unfavorable group displays increased expression of IRF4, dysregulated immune signaling and an immunosuppressive microenvironment. However, little is known on the molecular mechanisms whereby IRF4 controls the crosstalk between B and T cells. To this aim, we leveraged Cγ1cre;irf4fl/fl mice that yield conditional deletion of irf4 (cKO) in germinal center B cells (GCB). After 10 days of immunization, cKO GCB showed significant increase of HLA-DR and CD40 expression, a skewed LZ:DZ ratio towards increased DZ, and suppression of plasma cells compared to WT. Upon irf4 deletion, we also found a significant increase of total CD4 and TFH cells and a decrease in Treg cells. To test if this effect was antigen-dependent, we cocultured sorted GCB from NP-OVA-immunized cKO and WT mice with CSFE-labeled OVA-specific OT-II CD4 T cells at a ratio of 2:1 for 5 days. We observed increased antigen specific proliferation of CD4 T cells with cKO GCB compared to WT. We then performed seahorse assay and mass spectrometry of the same GCB to define if IRF4-dependent B cell metabolic reprogramming influences immune responses. GCB cKO had lower glycolytic ECAR and increased intracellular glucose and non-essential amino acids than WT cells, suggesting a metabolic shift from glycolysis to the TCA cycle. Thus, IRF4 controls immune and metabolic signaling with functional effects on the B-T cell crosstalk. Funding Sources NCI K08CA279652, LRF Clinical Investigator Award (CDA 1020588), the Gerstner Family Career Development Award, the ASH Junior Faculty Scholar Award. Topic Categories Tumor Immunology: Cellular Responses and Tumor Microevironment (TIME)

Abstract Description Background Castration-resistant prostate cancers (CRPC) do not respond to immunotherapy.1,2 Prostate cancer-derived immune factors (IF) like sPD-L1 drive immunosuppression and resistance.3 Circulating tumor-reactive T cells (TTR; CD8+CD11ahigh) and other immune subsets predict progression-free survival in oligometastatic CRPC after radiotherapy.4 A better understanding of the levels and changes in IF in CRPC—a tumor immune “circulome”—may lead to better immunotherapeutic approaches. Methods 79 men with oligometastatic castration-resistant prostate cancer were treated with radiation and blood was collected before and after treatment (NCT02816983). Peripheral blood mononuclear cells (PBMC) were subjected to immunophenotyping, and plasma was subjected to the Olink Explore panel assay measuring levels of 5,416 IFs using a proximity extension assay technology. Results High baseline soluble levels of PD-L1, PD-L2, TIM-3, and other selected immunotherapy-related IFs predicted inferior outcomes, including overall survival. Conversely, high soluble CTLA-4 levels strongly predicted superior outcomes (p < 0.0001). These results additionally correlated with changes in circulating tumor-reactive (TTR) T cells in these patients. Discussion IF influence systemic immunity in CRPC. These results suggest a multifactorial mechanism by which prostate cancers drive systemic immunosuppression and resistance to immunotherapy. Ongoing efforts to address IFs may help overcome CRPC immunotherapy resistance.5 Funding Sources Department of Defense HT9425-24-1-0622, An Atlas of the Tumor Immune Circulome (PI: Orme) Prostate Cancer Foundation Young Investigator Award (PI: Orme) National Cancer Institute K12 CA90628-23 (PI: Orme) Topic Categories Tumor Immunology: Checkpoints, Prevention, and Treatment (TIPT)

Abstract Description Signal regulatory protein-α (SIRPα) is a key member of the “do-not-eat-me” signaling pathway and is expressed on intratumoral monocytes/macrophages in lymphoma. In this study, we found that SIRPα is also expressed by a subset of malignant B cells in lymphoma. SIRPα+ B cells have a memory phenotype with germinal center (GC) dark-zone features, and RNA velocity analysis showed that SIRPα+ B cells exhibit higher spliced transcripts confirming a more mature phenotype. SIRPα expression is substantially upregulated by treating lymphoma B cells with LPS or CpG and SIRPα+ B cells showed increased proliferation when compared to SIRPα- B cells, which was associated with reduced cellular autophagy. SIRPα+ B cells exhibited increased expression of BAFF signaling pathway molecules as the likely mechanism of increased proliferation and survival. Furthermore, in co-culture, SIRPα+ lymphoma B cells suppressed T-cell function and blocking SIRPa/CD47 signaling reversed T cell inhibition. Clinically, we found that a greater number of SIRPα+ B cells present in the diagnostic tumor biopsies of FL patients was associated with a significantly poorer overall survival. We conclude that SIRPα+ lymphoma B cells are a proliferative component of indolent lymphomas, are associated with T-cell suppression, and increased prevalence of these cells is associated with a poor prognosis. Targeting SIRPα+ B cells in lymphoma may therefore be a promising therapeutic strategy. Topic Categories Tumor Immunology: Cellular Responses and Tumor Microevironment (TIME)