The challenging task of enumerating blasts in the bone marrow

Digital morphology compared to the optical microscope: A validation study on reporting bone marrow aspirates

INTRODUCTION

This study aims to evaluate the effectiveness and reliability of the utilization for clinical reporting of the evaluation of digital images of bone marrow aspirates by morphologists and their comparability with the classic microscopic morphological evaluation.

METHODS

We scanned 180 consecutive bone marrow needle aspirates smears using the "Metafer4 VSlide" whole slide imaging (WSI) digital scanning system. We evaluated the statistical comparability and the risk of bias of the microscopic readings with those performed on the screen on the digitized medullary images.

RESULTS

The evaluation of cellularity on the screen was equivalent, with a higher frequency of "normal" than the analysis of digital preparations. The means and medians of the percentage values obtained on the different cell populations with the microscopic and digital reading were comparable as the main categories are concerned, with an average difference equal to 0 for the neutrophilic and eosinophilic granulocytic series, at -0.2% for the total myeloid cells, at 1.2% for the erythroid series, at -0.4% for the lymphocytes and at -0.4% for the blasts. Dysplastic features were consistently identified in 69/71 cell lineages.

CONCLUSION

Our study demonstrated that screen evaluation of digitized bone marrow needle aspirates provides quantitative and qualitative results comparable to traditional microscopic analysis of the corresponding slide smears. Digital images offer significant benefits in reducing the workload of experienced operators, reproducibility and sharing of observations, and image preservation. Even in routine diagnostic activities, their use does not alter the quality of the results obtained in evaluating bone marrow needle aspirates.

Digital Microscopy Augmented by Artificial Intelligence to Interpret Bone Marrow Samples for Hematological Diseases

Analysis of bone marrow aspirates (BMAs) is an essential step in the diagnosis of hematological disorders. This analysis is usually performed based on a visual examination of samples under a conventional optical microscope, which involves a labor-intensive process, limited by clinical experience and subject to high observer variability. In this work, we present a comprehensive digital microscopy system that enables BMA analysis for cell type counting and differentiation in an efficient and objective manner. This system not only provides an accessible and simple method to digitize, store, and analyze BMA samples remotely but is also supported by an Artificial Intelligence (AI) pipeline that accelerates the differential cell counting process and reduces interobserver variability. It has been designed to integrate AI algorithms with the daily clinical routine and can be used in any regular hospital workflow.

Flow Cytometric Assessment of Myelodysplastic Syndromes/Neoplasms

Myelodysplastic syndromes/neoplasms (MDS) are a heterogeneous class of hematopoietic stem cell neoplasms characterized by ineffective hematopoiesis leading to peripheral cytopenias. This group of diseases is typically diagnosed using a combination of clinical, morphologic, and genetic criteria. Many studies have described the value of multiparametric flow cytometry (MFC) in the diagnosis, classification, and prognostication of MDS. This review summarizes the approach to MDS diagnosis and immunophenotypic characterization using MFC and describes the current state while highlighting future opportunities and potential pitfalls.

Diagnosis of erythroid dysplasia by flow cytometry: a review

INTRODUCTION

The diagnosis of myelodysplastic syndrome (MDS) is complex. Flow cytometric analysis of the myelomonocytic compartment can be helpful, but it is highly subjective and reproducibility by non-specialized groups is unclear. Analysis of the erythroid lineage by flow cytometry is emerging as potentially more reproducible and easier to conduct, while keeping a high diagnostic performance.

AREAS COVERED

We review the evidence in this area, including 1) the use of well-established markers - CD71 and CD36 - and other less well-established markers and parameters; 2) the use of flow cytometric scores for the erythroid lineage; and 3) additional aspects, including the emergence of computational tools and the roles of flow cytometry beyond diagnosis. Finally, we discuss the limitations with the current evidence, including 1) the impact of the sample processing protocol and reagents on the results, 2) the lack of a standard gating strategy, and 3) conceptualization and design issues in the available publications.

EXPERT OPINION

We end by offering our recommendations for the current use - and our personal take on the value - of the analysis of erythroid lineage by flow cytometry.

Digital assessment of peripheral blood and bone marrow aspirate smears

The diagnosis of benign and neoplastic hematologic disorders relies on analysis of peripheral blood and bone marrow aspirate smears. As demonstrated by the widespread laboratory adoption of hematology analyzers for automated assessment of peripheral blood, digital analysis of these samples provides many significant benefits compared to relying solely on manual review. Nonetheless, analogous instruments for digital bone marrow aspirate smear assessment have yet to be clinically implemented. In this review, we first provide a historical overview detailing the implementation of hematology analyzers for digital peripheral blood assessment in the clinical laboratory, including the improvements in accuracy, scope, and throughput of current instruments over prior generations. We also describe recent research in digital peripheral blood assessment, particularly in the development of advanced machine learning models that may soon be incorporated into commercial instruments. Next, we provide an overview of recent research in digital assessment of bone marrow aspirate smears and how these approaches could soon lead to development and clinical adoption of instrumentation for automated bone marrow aspirate smear analysis. Finally, we describe the relative advantages and provide our vision for the future of digital assessment of peripheral blood and bone marrow aspirate smears, including what improvements we can soon expect in the hematology laboratory.

The flow cytometry myeloid progenitor count: A reproducible parameter for diagnosis and prognosis of myelodysplastic syndromes

BACKGROUND

The bone marrow blast count is central to the diagnosis and monitoring of myelodysplastic syndromes (MDS). It is an independent risk factor for worse prognosis whether based on the morphology blast count or the flow cytometry (FC) myeloid progenitor (MyP) count. It is a principal population in FC MDS analysis also because once defined; it provides significant contributions to the overall FC MDS score.

METHODS

We elected to investigate inter-analyst agreement for the most fundamental parameter of the FC MDS diagnostic score: the MyP count. A common gating strategy was agreed and used by seven cytometrists for blind analysis of 34 routine bone marrows sent for MDS work-up. Additionally, we compared the results with a computational approach.

RESULTS

Concordance was excellent: Intraclass correlation was 0.993 whether measuring %MyP of total cells or CD45+ cells, and no significant difference was observed between files from different centers or for samples with abnormal MyP phenotypes. Computational and manual results were similar. Applying the common strategy to individual laboratories' control cohorts produced similar MyP reference ranges across centers.

CONCLUSION

The FC MyP count offers a reliable diagnostic and prognostic measurement in MDS. The use of manual and computational approaches side by side may allow for optimizing both strategies. Considering its known prognostic power, the MyP count could be considered a useful and reliable addition to existing prognostic scoring systems.

An Automated Pipeline for Differential Cell Counts on Whole-Slide Bone Marrow Aspirate Smears

The pathologic diagnosis of bone marrow disorders relies in part on the microscopic analysis of bone marrow aspirate (BMA) smears and the manual counting of marrow nucleated cells to obtain a differential cell count (DCC). This manual process has significant limitations, including the analysis of only a small subset of optimal slide areas and nucleated cells, as well as interobserver variability due to differences in cell selection and classification. To address these shortcomings, we developed an automated machine learning-based pipeline for obtaining 11-component DCCs on whole-slide BMAs. This pipeline uses a sequential process of identifying optimal BMA regions with high proportions of marrow nucleated cells, detecting individual cells within these optimal areas, and classifying these cells into 1 of 11 DCC components. Convolutional neural network models were trained on 396,048 BMA region, 28,914 cell boundary, and 1,510,976 cell class images from manual annotations. The resulting automated pipeline produced 11-component DCCs that demonstrated a high statistical and diagnostic concordance with manual DCCs among a heterogeneous group of testing BMA slides with varying pathologies and cellularities. Additionally, we demonstrated that an automated analysis can reduce the intraslide variance in DCCs by analyzing the whole slide and marrow nucleated cells within all optimal regions. Finally, the pipeline outputs of region classification, cell detection, and cell classification can be visualized using whole-slide image analysis software. This study demonstrates the feasibility of a fully automated pipeline for generating DCCs on scanned whole-slide BMA images, with the potential for improving the current standard of practice for utilizing BMA smears in the laboratory analysis of hematologic disorders.

Detection of Metastatic Tumor Cells in the Bone Marrow Aspirate Smears by Artificial Intelligence (AI)-Based Morphogo System

Introduction

Metastatic carcinomas of bone marrow (MCBM) are characterized as tumors of non-hematopoietic origin spreading to the bone marrow through blood or lymphatic circulation. The diagnosis is critical for tumor staging, treatment selection and prognostic risk stratification. However, the identification of metastatic carcinoma cells on bone marrow aspiration smears is technically challenging by conventional microscopic screening.

Objective

The aim of this study is to develop an automatic recognition system using deep learning algorithms applied to bone marrow cells image analysis. The system takes advantage of an artificial intelligence (AI)-based method in recognizing metastatic atypical cancer clusters and promoting rapid diagnosis.

Methods

We retrospectively reviewed metastatic non-hematopoietic malignancies in bone marrow aspirate smears collected from 60 cases of patients admitted to Zhongshan Hospital. High resolution digital bone marrow aspirate smear images were generated and automatically analyzed by Morphogo AI based system. Morphogo system was trained and validated using 20748 cell cluster images from randomly selected 50 MCBM patients. 5469 pre-classified cell cluster images from the remaining 10 MCBM patients were used to test the recognition performance between Morphogo and experienced pathologists.

Results

Morphogo exhibited a sensitivity of 56.6%, a specificity of 91.3%, and an accuracy of 82.2% in the recognition of metastatic cancer cells. Morphogo’s classification result was in general agreement with the conventional standard in the diagnosis of metastatic cancer clusters, with a Kappa value of 0.513. The test results between Morphogo and pathologists H1, H2 and H3 agreement demonstrated a reliability coefficient of 0.827. The area under the curve (AUC) for Morphogo to diagnose the cancer cell clusters was 0.865.

Conclusion

In patients with clinical history of cancer, the Morphogo system was validated as a useful screening tool in the identification of metastatic cancer cells in the bone marrow aspirate smears. It has potential clinical application in the diagnostic assessment of metastatic cancers for staging and in screening MCBM during morphology examination when the symptoms of the primary site are indolent.

WT1 Expression Levels Combined with Flow Cytometry Blast Counts for Risk Stratification of Acute Myeloid Leukemia and Myelodysplastic Syndromes

Wilm's tumor 1 (WT1), a zinc-finger transcription factor and an epigenetic modifier, is frequently overexpressed in several hematologic disorders and solid tumors, and it has been proposed as diagnostic and prognostic marker of acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS). However, the exact role of WT1 in leukemogenesis and disease progression remains unclear. In this real-world evidence retrospective study, we investigated prognostic role of WT1-mRNA expression levels in AML and MDS patients and correlations with complete blood counts, flow cytometry counts, and molecular features. A total of 71 patients (AML, n = 46; and MDS, n = 25) were included in this study, and WT1 levels were assessed at diagnosis, during treatment and follow-up. We showed that WT1 expression levels were inversely correlated with normal hemopoiesis in both AML and MDS, and positively associated with blast counts. Flow cytometry was more sensitive and specific in distinguishing normal myeloid cells from neoplastic counterpart even just using linear parameters and CD45 expression. Moreover, we showed that a simple integrated approach combining blast counts by flow cytometry, FLT3 mutational status, and WT1 expression levels might be a useful tool for a better prognostic definition in both AML and MDS patients.

Clinical significance of blast percentage assessed by bone marrow trephine biopsy and aspirate smear of myeloid malignancies

The blast percentage in bone marrow (BM) can be evaluated through biopsy and aspiration, which is essential for diagnosing myeloid neoplasms especially for dividing myelodysplastic syndrome (MDS)/acute myeloid leukaemia (AML). However, methods for integrating the results of biopsy and smear have yet to be developed, particularly for cases in which the results fall on both sides of the cut-off value (10% or 20%). We studied 188 cases of MDS/AML initially diagnosed during 2011-2015 by using concomitant BM biopsy and aspiration and used different methods to compare the estimated blast percentages. A linear relationship was noted between the blast percentages estimated through biopsy and smear (R²=0.765). When the blast percentage was classified into four relevant clinical categories (<5%, 5-9%, 10-19%, and ≥20%), the total concordance between the results of the biopsy and smear was 76.1%. Although the prognostic values obtained through biopsy and smear were not significantly different, using the higher blast percentage estimation by biopsy and smear fared better in classifying patients into categories of 10-19% and ≥20% and demonstrated survival significance in both univariate and multivariate analyses. Subgroup analyses demonstrated that BM blast percentages had no prognostic significance when patients underwent intensive chemotherapy. However, blast percentages of ≥10% indicated poor prognosis for patients receiving only supportive care. In conclusion, most of the clinically relevant categories of blast percentages estimated through concomitant BM biopsy and smear were concordant. When the categories were different, the best prognostic prediction method was to select the higher blast percentage determined through biopsy and smear to diagnose MDS/AML.

Case Report: Asymmetric Bone Marrow Involvement in Patients With Acute Leukemia After Allogeneic Hematopoietic Stem Cell Transplantation

After allogeneic hematopoietic stem cell transplantation (allo-HSCT), acute leukemia relapse is common, and asymmetric bone marrow recurrence hasn't been reported. Because the anatomical distribution of acute leukemia clones in the bone marrow after allo-HSCT is presumed to be diffuse, bone marrow aspirations are performed in single site. The first case was a 20-year-old man who was diagnosed with acute myelomonocytic leukemia and received haploidentical allo-HSCT. Routine bone marrow biopsy of his left posterior iliac bone marrow showed 52% leukemia blasts, while the right side had 0% blasts 10 days later. The second case was a 23-year-old woman who was diagnosed with acute B lymphoblastic leukemia and received HLA-identical sibling allo-HSCT. Although 62% of blasts were found in her left iliac marrow on day +122, 0% of blasts were found on a sample obtained from the right iliac crest on day +128. Bilateral iliac bone marrow pathology and whole-body ¹⁸F-FDG PET/CT scans confirmed that the leukemic infiltration in her bone marrow was asymmetric. To our knowledge, these are the first case reports of asymmetric bone marrow infiltration of blasts in acute leukemia patients after allo-HSCT. Bilateral posterior iliac crest aspirations or ¹⁸F-FDG-PET/CT scans may help distinguish such involvement.

A Hematologist-Level Deep Learning Algorithm (BMSNet) for Assessing the Morphologies of Single Nuclear Balls in Bone Marrow Smears: Algorithm Development

Background

Bone marrow aspiration and biopsy remain the gold standard for the diagnosis of hematological diseases despite the development of flow cytometry (FCM) and molecular and gene analyses. However, the interpretation of the results is laborious and operator dependent. Furthermore, the obtained results exhibit inter- and intravariations among specialists. Therefore, it is important to develop a more objective and automated analysis system. Several deep learning models have been developed and applied in medical image analysis but not in the field of hematological histology, especially for bone marrow smear applications.

Objective

The aim of this study was to develop a deep learning model (BMSNet) for assisting hematologists in the interpretation of bone marrow smears for faster diagnosis and disease monitoring.

Methods

From January 1, 2016, to December 31, 2018, 122 bone marrow smears were photographed and divided into a development cohort (N=42), a validation cohort (N=70), and a competition cohort (N=10). The development cohort included 17,319 annotated cells from 291 high-resolution photos. In total, 20 photos were taken for each patient in the validation cohort and the competition cohort. This study included eight annotation categories: erythroid, blasts, myeloid, lymphoid, plasma cells, monocyte, megakaryocyte, and unable to identify. BMSNet is a convolutional neural network with the YOLO v3 architecture, which detects and classifies single cells in a single model. Six visiting staff members participated in a human-machine competition, and the results from the FCM were regarded as the ground truth.

Results

In the development cohort, according to 6-fold cross-validation, the average precision of the bounding box prediction without consideration of the classification is 67.4%. After removing the bounding box prediction error, the precision and recall of BMSNet were similar to those of the hematologists in most categories. In detecting more than 5% of blasts in the validation cohort, the area under the curve (AUC) of BMSNet (0.948) was higher than the AUC of the hematologists (0.929) but lower than the AUC of the pathologists (0.985). In detecting more than 20% of blasts, the AUCs of the hematologists (0.981) and pathologists (0.980) were similar and were higher than the AUC of BMSNet (0.942). Further analysis showed that the performance difference could be attributed to the myelodysplastic syndrome cases. In the competition cohort, the mean value of the correlations between BMSNet and FCM was 0.960, and the mean values of the correlations between the visiting staff and FCM ranged between 0.952 and 0.990.

Conclusions

Our deep learning model can assist hematologists in interpreting bone marrow smears by facilitating and accelerating the detection of hematopoietic cells. However, a detailed morphological interpretation still requires trained hematologists.

Multiplexed single-cell morphometry for hematopathology diagnostics

The diagnosis of lymphomas and leukemias requires hematopathologists to integrate microscopically visible cellular morphology with antibody-identified cell surface molecule expression. To merge these into one high-throughput, highly multiplexed, single-cell assay, we quantify cell morphological features by their underlying, antibody-measurable molecular components, which empowers mass cytometers to 'see' like pathologists. When applied to 71 diverse clinical samples, single-cell morphometric profiling reveals robust and distinct patterns of 'morphometric' markers for each major cell type. Individually, lamin B1 highlights acute leukemias, lamin A/C helps distinguish normal from neoplastic mature T cells, and VAMP-7 recapitulates light-cytometric side scatter. Combined with machine learning, morphometric markers form intuitive visualizations of normal and neoplastic cellular distribution and differentiation. When recalibrated for myelomonocytic blast enumeration, this approach is superior to flow cytometry and comparable to expert microscopy, bypassing years of specialized training. The contextualization of traditional surface markers on independent morphometric frameworks permits more sensitive and automated diagnosis of complex hematopoietic diseases.

MRD evaluation of AML in clinical practice: are we there yet?

MRD technologies increase our ability to measure response in acute myeloid leukemia (AML) beyond the limitations of morphology. When applied in clinical trials, molecular and immunophenotypic MRD assays have improved prognostic precision, providing a strong rationale for their use to guide treatment, as well as to measure its effectiveness. Initiatives such as those from the European Leukemia Network now provide a collaborative knowledge-based framework for selection and implementation of MRD assays most appropriate for defined genetic subgroups. For patients with mutated-NPM1 AML, quantitative polymerase chain reaction (qPCR) monitoring of mutated-NPM1 transcripts postinduction and sequentially after treatment has emerged as a highly sensitive and specific tool to predict relapse and potential benefit from allogeneic transplant. Flow cytometric MRD after induction is prognostic across genetic risk groups and can identify those patients in the wild-type NPM1 intermediate AML subgroup with a very high risk for relapse. In parallel with these data, advances in genetic profiling have extended understanding of the etiology and the complex dynamic clonal nature of AML, as well as created the opportunity for MRD monitoring using next-generation sequencing (NGS). NGS AML MRD detection can stratify outcomes and has potential utility in the peri-allogeneic transplant setting. However, there remain challenges inherent in the NGS approach of multiplex quantification of mutations to track AML MRD. Although further development of this methodology, together with orthogonal testing, will clarify its relevance for routine clinical use, particularly for patients lacking a qPCR genetic target, established validated MRD assays can already provide information to direct clinical practice.

Statistics and measurable residual disease (MRD) testing: uses and abuses in hematopoietic cell transplantation

SERIES EDITORS' NOTE

The decision whether to recommend a transplant to someone with acute leukemia in first remission is complex and challenging. Diverse, often confounded co-variates interact to influence one's recommendation. Briefly, the decision metric can be viewed in three spheres: (1) subject-; (2) transplant-; and (3) disease-related co-variates. Subject-related co-variates include items such as age and comorbidities. Transplant-related co-variates include items such as donor-types, graft source, proposed conditioning and pre- and post-transplant immune suppression.But what of disease-related variables? Previously haematologists relied on co-variates such as WBC at diagnosis, chemotherapy cycles to achieve first remission, cytogenetics and most recently, mutation topography. However, these co-variates have largely been replaced by results of measurable residual disease (MRD)-testing. Many chemotherapy-only and transplant studies report strong correlations between results of MRD-testing on therapy outcomes such as cumulative incidence of relapse (CIR), leukemia-free survival (LFS) or survival. (CIR makes biological sense in a transplant context whereas LFS and survival do not give competing causes of death such as transplant-related mortality (TRM), graft-versus-host disease and interstitial pneumonia unrelated to relapse probability).This raises the question of how useful results are of MRD-testing in predicting CIR after transplants. Elsewhere we discussed accuracy and precision of MRD-testing in predicting outcomes of therapy of acute myeloid leukemia (Estey E, Gale RP. Leukemia 31:1255-1258, 2017; Hourigan CS, Gale RP, Gormley NJ, Ossenkoppele GJ, Walter RB. Leukemia 31:1482-1490, 2017). Briefly put, not terribly good. Although results of MRD-testing are often the most powerful predictor of CIR in multivariable analyses, the C-statistic (a measure of prediction accuracy) is often only about 0.75. This is much better than flipping a fair coin but far from ideal.In the typescript which follows, Othus and colleagues discuss statistical issues underlying MRD-testing in the context of haematopoietic cell transplants. We hope readers, especially haematologists who often need to make transplant recommendations to people with acute leukemia in first remission, will read it carefully and critically. The bottom line is MRD-test data are useful but considerable uncertainty is unavoidable with substantial false-positive and -negative rates. We need to acknowledge this uncertainty to ourselves and to the people we counsel. The authors quote Voltaire who said: Doubt is not a pleasant condition, but certainty is an absurd one. Sadly so, but we do the best we can. Robert Peter Gale, Imperial College London, and Mei-Jie Zhang, Medical College of Wisconsin and CIBMTR.