A Completely Digital Workflow for Differentials in Bone Marrow Cytomorphology Supported By Machine Learning Provides Promising Results in Object Detection

Background: Cytomorphology is an essential method to assess disease phenotypes. Recently, promising results of automation, digitalization and machine learning (ML) for this gold standard have been demonstrated. We reported on successful integration of such workflows into our lab routine, including automated scanning of peripheral blood smears and ML-based classification of blood cell images (ASH 2020). Following this pilot project, we are focusing on an equivalent approach for bone marrow. Aim: To establish a multistep-approach including scan of bone marrow smears and detection/classification of all kinds of bone marrow cell types in healthy individuals and leukemia patients. Methods: The method includes a pre-scan at 10x magnification for detecting suitable "areas of interest" (AOI) for cytomorphological analysis, a high resolution capture of a predefinable number of AOI at 40x magnification (always using oil) and an automated object detection and classification. For all scanning tasks, a Metafer Scanning System (Zeiss Axio Imager.Z2 microscope, automatic slide feeder SFx80 and automated oil disperser) from MetaSystems (Altlussheim, GER) was used. To generate training data for AOI detection, 37 bone marrow smears were scanned at 10x magnification. 6 different quality classes of regions (based on number and distribution of cells) were annotated by hem experts using polygons. In total, 185,000 grid images were extracted from the annotated regions and used for training a deep neural network (DNN) to distinguish the 6 quality classes and to generate a position list for a high resolution scan (40x magnification). In addition, we scanned the labeled AOI of 68 smears at 40x magnification, acquiring colour images (2048x1496 pixels) of bone marrow cell layers. Each single cell was labeled by human investigators using rectangular bounding boxes (in total: 47,118 cells in 511 images). We set up a supervised ML model, using the labeled 40x images as an input. We fine-tuned the COCO dataset pre-trained YOLOv5 model with our dataset and evaluated using 5-fold cross valuation. To reduce overfitting, image augmentation algorithms were applied. Results: Our first DNN was able to detect (10x magnification) and capture (40x magnification) AOI in bone marrow smears, sorted by quality and in acceptable time spans. Average time for the 10x pre-scan was 6 min. From the resulting position list, the 50 positions with highest quality values were acquired at an average of 1:30 min. Our second, independent DNN was able to detect nucleated cells at 94% sensitivity and 75% precision in unlabeled bone marrow images (40x magnification). In this model, we overweighted recall over precision (5:1) to avoid missing any objects of interest, assuming that false positive labels could be corrected by human investigators when reviewing digital images. For the classification of single cells, a third independent DNN will be necessary. Actually, different approaches are being tested, including our existing blood cell classifier and a former collaborative bone marrow classification model based on a training set of 100,000 annotated bone marrow cells. Depending on these results, new training data for generation of a completely new model could be assessed. The two existing models enable a fully automated digital workflow including scan of bone marrow smears and delivery of single cell image galleries for human classification already now. Conclusion: We here present solutions for multiple-DNN-based tools for bone marrow cytomorphology. They allow working digitally and remotely in routine diagnostics. Final solutions will offer single cell classifications and galleries for human review and include real time training of respective classifier models with dynamic datasets. Figure 1 Figure 1. Disclosures Haferlach: MLL Munich Leukemia Laboratory: Other: Part ownership. Kern: MLL Munich Leukemia Laboratory: Other: Part ownership. Haferlach: MLL Munich Leukemia Laboratory: Other: Part ownership.

Tissue-resident M2 macrophages directly contact primary sensory neurons in the sensory ganglia after nerve injury

Background Macrophages in the peripheral nervous system are key players in the repair of nerve tissue and the development of neuropathic pain due to peripheral nerve injury. However, there is a lack of information on the origin and morphological features of macrophages in sensory ganglia after peripheral nerve injury, unlike those in the brain and spinal cord. We analyzed the origin and morphological features of sensory ganglionic macrophages after nerve ligation or transection using wild-type mice and mice with bone-marrow cell transplants. Methods After protecting the head of C57BL/6J mice with lead caps, they were irradiated and transplanted with bone-marrow-derived cells from GFP transgenic mice. The infraorbital nerve of a branch of the trigeminal nerve of wild-type mice was ligated or the infraorbital nerve of GFP-positive bone-marrow-cell-transplanted mice was transected. After immunostaining the trigeminal ganglion, the structures of the ganglionic macrophages, neurons, and satellite glial cells were analyzed using two-dimensional or three-dimensional images. Results The number of damaged neurons in the trigeminal ganglion increased from day 1 after infraorbital nerve ligation. Ganglionic macrophages proliferated from days 3 to 5. Furthermore, the numbers of macrophages increased from days 3 to 15. Bone-marrow-derived macrophages increased on day 7 after the infraorbital nerve was transected in the trigeminal ganglion of GFP-positive bone-marrow-cell-transplanted mice but most of the ganglionic macrophages were composed of tissue-resident cells. On day 7 after infraorbital nerve ligation, ganglionic macrophages increased in volume, extended their processes between the neurons and satellite glial cells, and contacted these neurons. Most of the ganglionic macrophages showed an M2 phenotype when contact was observed, and little neuronal cell death occurred. Conclusion Most of the macrophages that appear after a nerve injury are tissue-resident, and these make direct contact with damaged neurons that act in a tissue-protective manner in the M2 phenotype. These results imply that tissue-resident macrophages signal to neurons directly through physical contact.

Diagnostic Value of Bone Marrow Cell Morphology in Visceral Leishmaniasis-Associated Hemophagocytic Syndrome of Two Cases

Background: Visceral leishmaniasis related-hemophagocytic lymphohistiocytosis (VL-HLH) is a hemophagocytic syndrome caused by Leishmania infection. VL-HLH is rare, especially in nonendemic areas where the disease is severe, and mortality rates are high. The key to diagnosing VL-HLH is to find the pathogen; therefore, the Leishmania must be accurately identified for timely clinical treatment.Case presentationWe retrospectively analyzed the clinical data, laboratory examination results and bone marrow cell morphology of two children with VL-HLH diagnosed via bone marrow cell morphology between July 2017 and January 2021 at Kunming Children’s Hospital of Yunnan, China.Two cases suspected of having malignant tumors at other hospitals and who had undergone ineffective long-term treatment were transferred to Kunming Children’s Hospital. They had repeated fevers, pancytopenia, hepatosplenomegaly, hypertriglyceridemia, and hypofibrinogenemia over a long period and met the HLH-2004 standard. Their HLH genetic test results were negative, and primary HLH was excluded. Both children underwent chemotherapy as per the HLH-2004 chemotherapy regimen , but it was ineffective and accompanied by serious infections. We found Leishmania amastigotes in their bone marrow via morphological examination of their bone marrow cells, which showed hemophagocytic cells; thus, the children were diagnosed with VL-HLH. After being transferred to a specialty hospital for treatment, the condition was well-controlled. Conclusion: Morphological examination of the bone marrow cells played an important role in diagnosing VL-HLH. When clinically diagnosing secondary HLH, VL-HLH should be considered in addition to common pathogens, especially in patients for whom HLH-2004 chemotherapy regimens are ineffective. For infants and young children, bone marrow cytology examinations should be performed several times and as early as possible to find the pathogens to reduce potential misdiagnoses.

THE IMPACT OF THE FOOD ADDITIVE E407A ON THE METABOLIC ACTIVITY OF DIFFERENT CELL CULTURES

Objectives. To study the effects of various concentrations of the food additive E407a (semi-refined carrageenan) on the metabolic activity of fetal liver cells, splenocytes, and bone marrow cells. Material and methods. Fetal liver, splenocytes and bone marrow cell cultures were incubated with the food additive E407a at concentrations varying from 0 mg/ml to 10 mg/ml for 24 hours (n=8). To analyze the effects of this food additive on the metabolic activity of cells, a colorimetric MTT assay was used. It is based on the ability of viable, metabolically active cells to convert 3 (4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide into formazan. The data were statistically processed using Kruskal-Wallis and Dunn’s criteria. Results. The bone marrow cell culture was found to be the most sensitive to carrageenan. More than a twofold statistically significant (p<0.0001) increase in the metabolic activity of bone marrow cells was observed when using E407a from 200 μg/ml and above. The metabolic activity of splenocytes increased approximately 1.5 times and over (p<0.0001) when using carrageenans at the concentration of 500 μg/ml and higher. Fetal liver cells turned out to be the most resistant to the direct toxic effect of the food additive E407a. Conclusions. The food additive E407a is cytotoxic to bone marrow cells and splenocytes at concentrations of 200 μg/ml and 500 μg/ml, respectively.

Tissue-resident M2 Macrophages Directly Contact Primary Sensory Neurons in the Sensory Ganglia after Nerve Injury

Background: Macrophages in the peripheral nervous system are key players in the repair of nerve tissue and the development of neuropathic pain due to peripheral nerve injury. However, there is a lack of information on the origin and morphological features of macrophages in sensory ganglia after peripheral nerve injury, unlike those in the brain and spinal cord. We analyzed the origin and morphological features of sensory ganglionic macrophages after nerve ligation or transection using wild-type mice and mice with bone-marrow cell transplants.Methods: After protecting the head of C57BL/6J mice with lead caps, they were irradiated and transplanted with bone-marrow-derived cells from GFP transgenic mice. The infraorbital nerve of a branch of the trigeminal nerve of wild-type mice was ligated or the infraorbital nerve of GFP-positive bone-marrow-cell-transplanted mice was transected. After immunostaining the trigeminal ganglia, the structures of the ganglionic macrophages, neurons, and satellite glial cells were analyzed using two-dimensional or three-dimensional images.Results: The number of damaged neurons in the trigeminal ganglia increased from day 1 after infraorbital nerve ligation. Ganglionic macrophages proliferated from days 3 to 5. Furthermore, the numbers of macrophages increased from days 3 to 15. Bone-marrow-derived macrophages increased on day 7 after the infraorbital nerve was transected in the trigeminal ganglia of GFP-positive bone-marrow-cell-transplanted mice but most of the ganglionic macrophages were composed of tissue-resident cells. On day 7 after infraorbital nerve ligation, ganglionic macrophages increased in volume, extended their processes between the neurons and satellite glial cells, and contacted these neurons. Most of the ganglionic macrophages showed an M2 phenotype when contact was observed, and little neuronal cell death occurred.Conclusions: Most of the macrophages that appear after a nerve injury are tissue-resident, and these make direct contact with damaged neurons that act in a tissue-protective manner in the M2 phenotype. These results imply that tissue-resident macrophages signal to neurons directly through physical contact.

Carbon Monoxide Suppresses Neointima Formation in Transplant Arteriosclerosis by Inhibiting Vascular Progenitor Cell Differentiation

Objective: Evidence indicates that bone marrow progenitor cells (BMPC) are a major contributor to neointima formation in transplant arteriosclerosis. HO-1 (heme oxygenase 1, Hmox1 ) and carbon monoxide (CO), a product of heme degradation by HO-1, ameliorate neointima formation by inhibiting proliferation of smooth muscle cells. We investigated the mechanism whereby HO-1 and CO modulate BMPC and mitigates neointima formation in transplant arteriosclerosis. Approach and Results: Using a murine model of aortic transplantation, bone marrow chimeric mice, and in vitro experiments, we report that CO does not inhibit mobilization of BMPC into the circulation or their homing to the vessel adventitia, but instead suppresses differentiation of BMPC into smooth muscle cells after they arrive in the adventitia. Specifically, the effect of CO on differentiation of BMPC into smooth muscle cell is mediated in part, by limiting PDGFR-β (platelet derived growth factor receptor-β) signaling. Hmox1 −/− BMPC exhibit a greater propensity to differentiate into smooth muscle cell in vitro, in part by regulating PDGFR-β + expression. Furthermore, wild-type mice transplanted with Hmox1 −/− bone marrow cells show augmented neointima formation after allografting versus control. CO exposure significantly ameliorated neointima formation, which remains more severe with Hmox1 −/− bone marrow cell versus air-treated mice receiving HO-1-expressing bone marrow cell, highlighting the importance of endogenous HO-1 in neointima formation. Conclusions: Host BMPC contribute to neointima formation in transplant arteriosclerosis and the protective effect afforded by HO-1/CO against neointima formation is mediated in part through the regulation of PDGFR-β expression. We propose that suppressing differentiation of BMPC is a major mechanism by which HO-1 and CO prevent neointima expansion after transplant.

Response to Comment on “Tumor-initiating cells establish an IL-33–TGF-β niche signaling loop to promote cancer progression”

Kamphuis et al. argue that macrophages accumulated in the proximity of tumor-initiating cells do not express the high-affinity immunoglobulin E receptor FcεRIα. Although we cannot exclude the possibility of nonspecific binding of anti-FcεRIα antibody (clone MAR-1), we provide evidence that macrophages in squamous cell carcinomas express FcεRIα and that IL-33 induces FcεRIα expression in bone marrow cell–derived macrophages.