Faculty Bibliography

The diagnosis of myelodysplastic neoplasms (MDS) requires examination of the bone marrow for morphologic evidence of dysplasia. We sought to determine whether a self-supervised learning (SSL) artificial intelligence-based image analysis approach can be used to reliably distinguish MDS from its clinically relevant mimics using bone marrow biopsies (BMBx). The whole-slide images (WSIs) of hematoxylin and eosin (H&E)- and reticulin-stained BMBx sections from 243 unique patients (89 MDS, 55 non-MDS cytopenic controls [NMCCs], and 99 negative control [NC] cases) were partitioned into image tiles for analysis. These image tiles were then processed using the Barlow Twins SSL model to identify histomorphologic phenotype clusters (HPCs). Review of the HPCs revealed the clusters enriched in MDS cases captured known histopathologic features of the disease, including hypercellularity (characterized by enrichment in hypercellular image tiles), dysplastic and loosely clustered megakaryocytes, increased immature hematopoietic cells, increased vascularity, fibrosis, and cell streaming patterns. For external validation, 95 MDS BMBx WSIs from an independent institution were analyzed using the trained model. The model demonstrated consistent HPC enrichment patterns, supporting its robustness and generalizability. The trained ensemble model using H&E- and reticulin-stained slides distinguished MDS from NCs with an area under the curve (AUC) of 0.82, and from age-matched NMCCs with an AUC of 0.80. These findings demonstrate the potential of SSL approaches to capture diagnostically relevant morphologic patterns and to improve the reproducibility of MDS diagnosis.

Cytology cell block specimens are essential for diagnosing patients with advanced-stage malignancy and often represents the only available strategy for therapeutic biomarker evaluation. The use of cell blocks preserves tumor cells, captures high-grade or metastatic populations, and retains meaningful microenvironmental context, making them well suited for IHC analysis. With the rapid expansion of computational pathology, deep learning-assisted biomarker interpretation in cell blocks is emerging as a promising frontier for improving reproducibility, reducing interobserver variability, and enabling quantitative assessment of spatial tumor-immune interactions. Because many treatment-defining biomarkers are routinely assessed on cytology cell blocks, this mini-review highlights artificial intelligence-based applications for PD-L1, HER2, ER/PR, Ki-67, ALK/ROS1, BRAF V600E, and p16 markers that directly inform decisions about immunotherapy, targeted therapy, and hormone therapy. Also reviewed are emerging predictive models that convey biomarker status directly from morphology, extending the utility of artificial intelligence beyond conventional IHC interpretation. Finally, spatial pathobiology-related opportunities afforded by cell block preparations are discussed, and future directions for integrating artificial intelligence-enabled analysis into cytology workflows are outlined. Together, these advances position cytology cell blocks as an important platform for computational biomarker interpretation and morphology-driven precision oncology.

The assessment of measurable residual disease (MRD) plays a critical role in acute myeloid leukemia (AML) treatment response evaluation and prognosis. However, current AML MRD detection by flow cytometry (FC) is limited in sensitivity due to immunophenotypic variability, similarities to normal hematopoietic stem/progenitor cells, and the lack of stable leukemia-associated immunophenotypes. A significant proportion of AML patients classified as MRD-negative by FC eventually relapse, likely due to the persistence of therapy-resistant leukemic stem cells (LSCs) that are not sensitively detected by routine clinical flow panels. Flow cytometry panels designed to detect LSC antigens, while promising, face challenges like immunophenotypic heterogeneity across AML subtypes, lack of standardized marker panels across laboratories, and limited validation. Here, we summarize the current state of FC-based LSC detection in AML, discussing commonly used markers, immunophenotypic variability, assay setup challenges, and we review recent clinical studies on LSC assessment, outlining their main findings and implications for prognosis and MRD integration. We also consider advances in spectral flow cytometry for improved LSC detection.

Adhesion G protein-coupled receptors (aGPCRs) are key cell-adhesion molecules involved in many cellular functions and contribute to human diseases, including cancer. aGPCRs are characterized by large extracellular regions that could serve as readily accessible antigens. However, the potential of aGPCRs as targets for biologic therapeutics has not been extensively explored. CD97, also known as ADGRE5, is an aGPCR that is upregulated in various cancer types, including acute myeloid leukemia (AML) and glioblastoma (GBM), and their respective cancer stem cells. Here, we developed antibody-drug conjugates (ADCs) targeting CD97 and assessed their efficacy against AML and GBM cells. We generated a panel of synthetic human antibodies targeting distinct epitopes of CD97, from which we identified an antibody that was efficiently internalized. This antibody binds to all isoforms of human CD97 but not to its close homolog, EMR2. Structure determination by single-particle cryo-electron microscopy revealed that this antibody targets the CD97 GPCR autoproteolysis-inducing (GAIN) domain, whose presence is conserved in aGPCRs, through an unconventional binding mode where it extensively utilizes the light chain framework for antigen recognition. Screening of conjugation methods and payloads resulted in a stable ADC that effectively killed AML and GBM cell lines, as well as patient-derived GBM stem cells, with minimal cytotoxicity against peripheral blood mononuclear cells from healthy donors. Our study demonstrates the therapeutic potential of targeting CD97, as well as the aGPCR GAIN domain in general, and uncovers a previously unrecognized surface that an antibody can utilize for antigen recognition.

Blood production is sustained by hematopoietic stem cells (HSCs), which are typically the only blood cells capable of long-term self-renewal. HSCs exhibit and depend on low levels of protein synthesis to self-renew. However, the mechanisms by which HSCs regulate protein synthesis to maintain their self-renewal capacity during proliferative stress and leukemogenesis remain unknown. Here we show CD99, a protein upregulated in leukemia stem cells (LSCs) in acute myeloid leukemia (AML), is required for the self-renewal of proliferating HSCs and LSCs. We found that loss of CD99 in HSCs and LSCs leads to increased protein synthesis and that their self-renewal capacity can be restored by translation inhibition. These data demonstrate a functional role for CD99 in constraining protein synthesis, which may promote the clonal expansion of HSCs and LSCs that leads to AML. Furthermore, these studies demonstrate that similar to HSCs, LSCs depend on maintenance of tightly regulated protein synthesis rates.

Hematopoietic stem cells (HSC) with multilineage potential are critical for T cell reconstitution after allogeneic hematopoietic cell transplantation (allo-HCT). The Kit^lo HSC subset is enriched for multipotential precursors, but their T cell potential remains poorly characterized. Using a preclinical allo-HCT mouse model, we demonstrate that Kit^lo HSCs provide superior thymic recovery and T cell reconstitution, resulting in improved immune responses to post-transplant infection. Kit^lo HSCs with augmented bone marrow (BM) lymphopoiesis mitigate age-associated thymic alterations and enhance T cell recovery in middle-aged mice. Mechanistically, chromatin profiling reveals Kit^lo HSCs exhibiting higher activity of lymphoid-specifying transcription factors, such as, ZBTB1. Zbtb1 deletion diminishes HSC engraftment and T cell potential; by contrast, reinstating Zbtb1 in megakaryocytic-biased Kit^hi HSCs rescues hematopoietic engraftment and T cell potential in vitro and in vivo. Furthermore, age-associated decline in Kit^lo HSCs is associated with diminished T lymphopoietic potential in aged BM precursors; meanwhile, Kit^lo HSCs in aged mice maintain enhanced lymphoid potential, but their per-cell capacity is diminished. Lastly, we observe an analogous human BM KIT^lo HSC subset with enhanced lymphoid potential. Our results thus uncover an age-related epigenetic regulation of lymphoid-competent Kit^lo HSCs for T cell reconstitution.

To determine the cellular and molecular basis of Castleman Disease (CD), we analyze the spatial proteome and transcriptome from a discovery (n = 9 cases) and validation (n = 13 cases) cohort of Unicentric CD, idiopathic Multicentric CD, HHV8-associated MCD, and reactive lymph nodes. CD shows increased stromal cells that form unique microenvironments. Interaction of activated follicular dendritic cell (FDC) cytoplasmic meshworks with mantle-zone B cells is associated with B-cell activation and differentiation. CXCL13+ FDCs, PDGFRA + T-zone reticular cells (TRC), and ACTA2-positive perivascular reticular cells (PRC) were the predominant source of increased VEGF expression and IL-6 signaling. MCD is characterized by increased TRC while UCD shows increased B-reticular cells (BRC). VEGF expression by FDCs is associated with peri-follicular neovascularization. FDC, TRC and PRC of CD activates JAK-STAT, TGFβ, and MAPK pathways via specific ligand-receptor interactions. Here, we show that stromal-cell activation and associated B cell activation and differentiation, neovascularization and stromal remodeling underlie CD.

BACKGROUND/UNASSIGNED:A thymoma is a tumor originating from thymic epithelial cells variably associated with non-neoplastic lymphocytes. T-lymphoblastic leukemia/lymphoma (T-LBL) is thought to arise from precursor T-cells from bone marrow-derived hematopoietic stem cells that migrate to the thymus. While the association of secondary hematopoietic malignancies in thymoma is well established, only rarely in the literature have T-LBL and thymoma been seen in association and the relationship is poorly understood. Occasionally, distinction between the two can be difficult as immature lymphocytes in thymoma resemble T-LBL both morphologically and immunophenotypically. An accurate diagnosis is essential as treatments vary between these two entities. CASE DESCRIPTION/UNASSIGNED:We present the interesting case of a 64-year-old male, former smoker, originally from Uzbekistan, with a mediastinal mass diagnosed as small cell carcinoma in his home country and treated with chemotherapy. After immigrating to the United States, a positron emission tomography (PET) scan demonstrated a large, metabolically active mediastinal mass. He presented to our institution where a biopsy with histomorphologic and immunohistochemical analysis was diagnostic of type B1 thymoma. He was lost to follow-up, but represented months later with B symptoms. Flow cytometry, cytogenetics, and bone marrow biopsy were diagnostic of T-LBL. Although he was started on chemotherapy, his disease progressed and he expired 6 months after initial presentation. Post-mortem analysis of the mediastinal mass revealed the co-occurrence of benign thymocytes and neoplastic T-LBL lymphoblasts, further confirmed as two distinct entities by T-cell receptor (TCR) sequencing. CONCLUSIONS/UNASSIGNED:Co-occurrence of thymoma and T-LBL is a well-documented, though poorly understood, phenomenon. Literature review for this phenomenon reveals that type B thymoma is most commonly associated with T-LBL in these co-occurrences. Most cases are diagnosed synchronously, though in metachronous cases, the diagnosis of thymoma has always preceded the diagnosis of T-LBL. Of note, recently developed LMO2 immunohistochemical stain is positive in malignant lymphoblasts but negative in benign thymocytes, allowing for post-mortem evaluation of this case to be determined as a synchronous presentation. These entities are difficult to distinguish and require a multimodal diagnostic approach including histology, immunohistochemistry, flow cytometry, cytogenetics, and TCR sequencing.

Acute myeloid leukemia (AML) is a complex cancer, yet advances in recent years from integrated genomics methods have helped improve diagnosis, treatment, and means of patient stratification. A recent example of a powerful, multimodal method is DOGMA-seq, which can measure chromatin accessibility, gene expression, and cell-surface protein levels from the same individual cell simultaneously. Previous bimodal single-cell techniques, such as CITE-seq (Cellular indexing of transcriptomes and epitopes), have only permitted the transcriptome and cell-surface protein expression measurement. DOGMA-seq, however, builds on this foundation and has implications for examining epigenomic, transcriptomic, and proteomic interactions between various cell types. This technique has the potential to be particularly useful in the study of cancers such as AML. This is because the cellular mechanisms that drive AML are rather heterogeneous and require a more complete understanding of the interplay between the genetic mutations, disruptions in RNA transcription and translation, and surface protein expression that cause these cancers to develop and evolve. This technique will hopefully contribute to a more clear and complete understanding of the growth and progression of complex cancers.

BRAF V600E mutations are frequently found in histiocytic/dendritic cell neoplasms such as Erdheim-Chester disease (ECD) and Langerhans cell histiocytosis (LCH), but few reports have also described BRAF mutations in Rosai-Dorfman disease (RDD), and even these cases may predominantly represent mixed histiocytosis. BRAF mutations have been studied in histiocytic/dendritic cell neoplasms and described to be associated with increased risk of relapse and long-term consequences, but few studies have examined BRAF V600E mutation in RDD, which is recognized as a neoplasm given the high frequency of MAPK pathway alterations. Here, we report a case of BRAF V600E-mutated RDD in a patient who presented with generalized lymphadenopathy. During our evaluation of this patient, we also found expression of PD-L1 in neoplastic histiocytes. During our review period, only few cases of RDD reported to harbor BRAF mutation or were evaluated for the expression of PDL1 by neoplastic cells. Given the potential challenges in distinguishing RDD from other histiocytic/dendritic cell neoplasms, including mixed histiocytosis with similar clinicopathological manifestations, we will discuss the current state of knowledge regarding the frequency and clinical impact of BRAF V600E in RDD, as well as the role of BRAF mutations in RDD pathogenesis. Distinction of BRAF V600E mutated histiocytic/dendritic cell neoplasms requires consideration of distinctive histopathological and immunophenotypic findings in appropriate clinical and radiologic setting. Given the increasing use of BRAF inhibitors as well as checkpoint blockade inhibitors to treat a number of cancers, we will discuss the clinical implications of the presence of BRAF V600E mutation and PD-L1 expression in RDD.