MAGIC-DR: An interpretable machine-learning guided approach for acute myeloid leukemia measurable residual disease analysis.

Multiparameter flow cytometry is widely used for acute myeloid leukemia minimal residual disease testing (AML MRD) but is time consuming and demands substantial expertise. Machine learning offers potential advancements in accuracy and efficiency, but has yet to be widely adopted for this application. To explore this, we trained single cell XGBoost classifiers from 98 diagnostic AML cell populations and 30 MRD negative samples. Performance was assessed by cross-validation. Predictions were integrated with UMAP as a heatmap parameter for an augmented/interactive AML MRD analysis framework, which was benchmarked against traditional MRD analysis for 25 test cases. The results showed that XGBoost achieved a median AUC of 0.97, effectively distinguishing diverse AML cell populations from normal cells. When integrated with UMAP, the classifiers highlighted MRD populations against the background of normal events. Our pipeline, MAGIC-DR, incorporated classifier predictions and UMAP into flow cytometry standard (FCS) files. This enabled a human-in-the-loop machine learning guided MRD workflow. Validation against conventional analysis for 25 MRD samples showed 100% concordance in myeloid blast detection, with MAGIC-DR also identifying several immature monocytic populations not readily found by conventional analysis. In conclusion, Integrating a supervised classifier with unsupervised dimension reduction offers a robust method for AML MRD analysis that can be seamlessly integrated into conventional workflows. Our approach can support and augment human analysis by highlighting abnormal populations that can be gated on for quantification and further assessment. This has the potential to speed up MRD analysis, and potentially improve detection sensitivity for certain AML immunophenotypes.

Causes of hypereosinophilia in 100 consecutive patients.

BACKGROUND: Hypereosinophilia (HE, persistent peripheral blood eosinophilia > 1.5 × 109 /L) and hypereosinophilic syndrome (HES, HE with end-organ damage) are classified as primary (due to a myeloid clone), secondary (due to a wide variety of reactive causes), or idiopathic. Diagnostic evaluation of eosinophilia is challenging, in part because secondary causes of HE/HES such as lymphocyte-variant HES (L-HES) and vasculitis are difficult to diagnose, and emerging causes such as immunoglobulin G4-related disease (IgG4-RD) have rarely been examined. OBJECTIVE AND METHODS: We reviewed 100 consecutive patients with HE/HES who underwent extensive evaluation for primary and secondary eosinophilia at a single tertiary care center to determine causes of HE/HES in a modern context. RESULTS: Six patients had primary HE/HES, 80 had a discrete secondary cause identified, and 14 had idiopathic HE/HES. The most common causes of secondary eosinophilia were L-HES/HES of unknown significance (L-HESus) (20), IgG4-RD (9), and eosinophilic granulomatosis with polyangiitis (EGPA) (8). CONCLUSIONS: In contrast to other large published series of HE/HES, most patients in this study were found to have a discrete secondary cause of eosinophilia and only 14 were deemed idiopathic. These findings highlight the importance of extensive evaluation for secondary causes of eosinophilia such as L-HES, IgG4-RD, and EGPA.

Post-Transfusion Hemophagocytosis Without Hemophagocytic Lymphohistiocytosis.

Hemophagocytosis refers to ingestion of hematopoietic elements or mature blood cells by another cell, typically by cells conventionally associated with phagocytic capacity. Although the finding of hemophagocytosis as a prominent feature in a patient's bone marrow might prompt consideration of a hemophagocytic syndrome (HPS) such as hemophagocytic lymphohistiocytosis (HLH) in a clinician's or pathologist's differential diagnosis, this morphologic feature can be nonspecific in the absence of other clinical and laboratory features of pathologic immune activation, which is the sine qua non of HPS/HLH. We describe three patients whose clinical presentations included transfusion-dependent anemia and whose bone marrow aspirates showed unexpectedly brisk hemophagocytosis of mature red blood cells. Despite striking morphologic hemophagocytosis, no patient met criteria for diagnosis of an HPS. Transfusion-associated hemophagocytosis and hyperferritinemia must be carefully distinguished from HLH through clinical and laboratory assessment. Biomarkers of pathologic immune activation are important diagnostic aids.

Therapeutic cells via functional modification: influence of molecular properties of polymer grafts on in vivo circulation, clearance, immunogenicity, and antigen protection.

Modulation of cell surface properties via functional modification is of great interest in cell-based therapies, drug delivery, and in transfusion. We study the in vivo circulation, immuogenicity, and mechanism of clearance of hyperbranched polyglycerol (HPG)-modified red blood cells (RBCs) as a function of molecular properties of HPGs. The circulation half-life of modified cells can be modulated by controlling the polymer graft concentration on RBCs; low graft concentrations (0.25 and 0.5 mM) showed normal circulation as that of control RBCs. Molecular weight of HPG did not affect the circulation of modified RBCs. HPG grafting on RBCs reduced CD47 self-protein accessibility in a graft concentration-dependent fashion. HPG-grafted RBCs are not immunogenic, as is evident from their similar circulation profile upon repeated administration in mice and monitoring over 100 days. Histological examination of the spleen, liver, and kidneys of the mice injected with modified RBCs revealed distinct differences, such as elevated iron deposits and an increase in the number of CD45 expressing cells at high graft concentration of HPGs (1.5 mM); no changes were seen at low graft concentration. The absence of iron deposits in the white pulp region of the spleen and its presence in the red pulp region indicates that the clearance of functional RBCs occurs in the venous sinuses mechanical filtering system, similar to the clearance of unmodified senescent RBCs. HPG modification at grafting concentrations that yield long circulation in mice produced camouflage of a large number of minor blood group antigens on human RBCs, demonstrating its utility in chronic transfusion. The normal circulation, nonimmunogenic nature, and the potential to modulate the circulation time of modified cells without toxicity make this HPG-based cell surface modification approach attractive for drug delivery and other cell-based therapies.

Influence of polymer architecture on antigens camouflage, CD47 protection and complement mediated lysis of surface grafted red blood cells.

Hyperbranched polyglycerol (HPG) and polyethylene glycol (PEG) polymers with similar hydrodynamic sizes in solution were grafted to red blood cells (RBCs) to investigate the impact of polymer architecture on the cell structure and function. The hydrodynamic sizes of polymers were calculated from the diffusion coefficients measured by pulsed field gradient NMR. The hydration of the HPG and PEG was determined by differential scanning calorimetry analyses. RBCs grafted with linear PEG had different properties compared to the compact HPG grafted RBCs. HPG grafted RBCs showed much higher electrophoretic mobility values than PEG grafted RBCs at similar grafting concentrations and hydrodynamic sizes indicating differences in the structure of the polymer exclusion layer on the cell surface. PEG grafting impacted the deformation properties of the membrane to a greater degree than HPG. The complement mediated lysis of the grafted RBCs was dependent on the type of polymer, grafting concentration and molecular size of grafted chains. At higher molecular weights and graft concentrations both HPG and PEG triggered complement activation. The magnitude of activation was higher with HPG possibly due to the presence of many hydroxyl groups per molecule. HPG grafted RBCs showed significantly higher levels of CD47 self-protein accessibility than PEG grafted RBCs at all grafting concentrations and molecular sizes. PEG grafted polymers provided, in general, a better shielding and protection to ABO and minor antigens from antibody recognition than HPG polymers, however, the compact HPGs provided greater protection of certain antigens on the RBC surface. Our data showed that HPG 20 kDa and HPG 60 kDa grafted RBCs exhibited properties that are more comparable to the native RBC than PEG 5 kDa and PEG 10 kDa grafted RBCs of comparable hydrodynamic sizes. The study shows that small compact polymers such as HPG 20 kDa have a greater potential in the generation of functional RBC for therapeutic delivery applications. The intermediate sized polymers (PEG or HPG) which showed greater antigen camouflage at lower grafting concentrations have significant potential in transfusion as universal red blood donor cells.