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.

Machine-based detection and classification for bone marrow aspirate differential counts: initial development focusing on nonneoplastic cells.

Bone marrow aspirate (BMA) differential cell counts (DCCs) are critical for the classification of hematologic disorders. While manual counts are considered the gold standard, they are labor intensive, time consuming, and subject to bias. A reliable automated counter has yet to be developed, largely due to the inherent complexity of bone marrow specimens. Digital pathology imaging coupled with machine learning algorithms represents a highly promising emerging technology for this purpose. Yet, training datasets for BMA cellular constituents, critical for building and validating machine learning algorithms, are lacking. Herein, we report our experience creating and employing such datasets to develop a machine learning algorithm to detect and classify BMA cells. Utilizing a web-based system that we developed for annotating and managing digital pathology images, over 10,000 cells from scanned whole slide images of BMA smears were manually annotated, including all classes that comprise the standard clinical DCC. We implemented a two-stage, detection and classification approach that allows design flexibility and improved classification accuracy. In a sixfold cross-validation, our algorithms achieved high overall accuracy in detection (0.959 ± 0.008 precision-recall AUC) and classification (0.982 ± 0.03 ROC AUC) using nonneoplastic samples. Testing on a small set of acute myeloid leukemia and multiple myeloma samples demonstrated similar detection and classification performance. In summary, our algorithms showed promising early results and represent an important initial step in the effort to devise a reliable, objective method to automate DCCs. With further development to include formal clinical validation, such a system has the potential to assist in disease diagnosis and prognosis, and significantly impact clinical practice.

Plasmablastic transformation of chronic lymphocytic leukemia: a review of literature and report on 2 cases.

Chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL/SLL) is the most common leukemia in adults in Western countries. Transformation of CLL/SLL to plasmablastic lymphoma (PBL) is exceedingly rare and often has an extremely poor response to treatment. A thorough molecular workup may help in determining clonality-relatedness and prognosis. We describe two cases of CLL/SLL that transformed into PBL, with an extensive molecular workup in one case, and a review of the literature.

Digital image analysis of erythroblastic islands in myelodysplastic syndromes.

INTRODUCTION: Myelodysplastic syndromes (MDS) encompass a diverse group of myeloid neoplasms for which the diagnosis of low-grade subtypes remains challenging. Erythroblastic islands (EBIs) are highly organized units of erythroid proliferation, differentiation, and enucleation. EBI disruption is frequently observed and is believed to be one of the early changes in MDS. METHODS: In this study, we digitally analyzed bone marrow biopsies dual stained with alpha-hemoglobin stabilizing protein (AHSP) and CD163 to quantitatively study features of EBIs in MDS, among MDS subtypes, as well as those in normal marrows and marrows with other causes of anemia. RESULTS: EBIs in MDS specimens were smaller in size and higher in density compared to both normal and non-MDS anemia specimens. Increased CD163 expression within the EBIs is observed in both MDS and other causes of anemia. A combination of increased EBI density and CD163 expression is seen in association with MDS with high-risk cytogenetics and multiple adverse mutations. CONCLUSION: As a proof-of-concept study, we show that EBI features can be relatively easily quantified with AHSP/CD163 dual immunohistochemistry and open-source imaging analysis software, highlighting those that are unique to MDS, and which may be prognostically relevant. Further studies of the measurable EBI features may provide valuable and novel tools to aid MDS diagnosis and prognostication in the era of digital pathology.

Haematology laboratory parameters to assess efficacy of CD19-, CD22-, CD33-, and CD123-directed chimeric antigen receptor T-cell therapy in haematological malignancies.

INTRODUCTION: Chimeric antigen receptor (CAR) T cell products are available to treat relapsed/refractory B-lymphoblastic leukaemia/lymphoma (B-ALL), diffuse large B-cell lymphoma, mantle-cell lymphoma, and myeloma. CAR products vary by their target epitope and constituent molecules. Hence, there are no common laboratory assays to assess CAR T cell expansion in the clinical setting. We investigated the utility of common haematology laboratory parameters to measure CAR T cell expansion and response. METHODS: Archived CellaVision images, absolute lymphocyte counts, and Sysmex CPD parameters spanning 1 month after CD19-CAR, UCAR19, CD22-CAR, CD33-CAR, and UCAR123 therapy were compared against donor lymphocyte infused control patients. Additionally, CellaVision images gathered during acute EBV infection were analysed. RESULTS: CellaVision images revealed a distinct sequence of three lymphocyte morphologies, common among CD19-CAR, CD22-CAR and UCAR19. This lymphocyte sequence was notably absent in CAR T cell non-responders and stem-cell transplantation controls, but shared some features seen during acute EBV infection. CD19-CAR engraftment kinetics monitored by quantitative PCR show an expansion and persistence phase and mirror CD19-CAR ALC kinetics. We show other novel CAR T cell therapies (UCAR19, CD22-CAR, CD33-CAR and UCAR123) display similar ALC expansion in responders and diminished ALC expansion in non-responders. Furthermore, the CPD parameter LY_WY fluorescence increased within the first week after CD19-CAR infusion, preceding the peak absolute lymphocyte count (ALC) by 3.7 days. CONCLUSION: Autologous and allogeneic CAR T cell therapy produce unique changes in common haematology laboratory parameters and could be a useful surrogate to follow CAR T-cell expansion after infusion.

CD30 + Primary intestinal T-cell lymphoma (unclassified) masquerading as chronic inflammation: a case report.

BACKGROUND: Primary intestinal T-cell lymphomas are uncommon malignancies that pose a diagnostic dilemma, because the clinical features and imaging findings commonly overlap with those encountered in inflammatory bowel diseases. CASE PRESENTATION: The current clinical case report describes the clinical history, laboratory findings and histopathological analysis from a patient with non-specific gastrointestinal symptoms with a presumptive clinical diagnosis of inflammatory bowel disease, and two intestinal biopsy specimens with non-specific findings. Due to the persistent symptoms a third biopsy was consistent with primary intestinal T-cell lymphoma, a diagnosis that was elusive for months after the initial presentation. Clinical correlation with laboratory and histopathological findings is required to establish a definitive diagnosis and to further stratify the patients. In addition, the neoplastic cells featured partial expression of CD30, which had relevant therapeutic implications. CONCLUSIONS: Suspicion for an intestinal T-cell lymphoproliferative disorder should always exist in patients with persistent abdominal symptoms with no clear etiology. The current discussion provides a summary and review of the key diagnostic histological features for the classification of primary intestinal T-cell lymphomas. In addition, the discussion describes how specific the histological findings are relevant for the clinical management decisions.

NuCLS: A scalable crowdsourcing approach and dataset for nucleus classification and segmentation in breast cancer.

BACKGROUND: Deep learning enables accurate high-resolution mapping of cells and tissue structures that can serve as the foundation of interpretable machine-learning models for computational pathology. However, generating adequate labels for these structures is a critical barrier, given the time and effort required from pathologists. RESULTS: This article describes a novel collaborative framework for engaging crowds of medical students and pathologists to produce quality labels for cell nuclei. We used this approach to produce the NuCLS dataset, containing >220,000 annotations of cell nuclei in breast cancers. This builds on prior work labeling tissue regions to produce an integrated tissue region- and cell-level annotation dataset for training that is the largest such resource for multi-scale analysis of breast cancer histology. This article presents data and analysis results for single and multi-rater annotations from both non-experts and pathologists. We present a novel workflow that uses algorithmic suggestions to collect accurate segmentation data without the need for laborious manual tracing of nuclei. Our results indicate that even noisy algorithmic suggestions do not adversely affect pathologist accuracy and can help non-experts improve annotation quality. We also present a new approach for inferring truth from multiple raters and show that non-experts can produce accurate annotations for visually distinctive classes. CONCLUSIONS: This study is the most extensive systematic exploration of the large-scale use of wisdom-of-the-crowd approaches to generate data for computational pathology applications.

Automated RNA In Situ Hybridization for 18 High Risk Human Papilloma Viruses in Squamous Cell Carcinoma of the Head and Neck: Comparison With p16 Immunohistochemistry.

Detection of human papilloma virus (HPV)-related head and neck squamous cell carcinoma (HNSCC) is important, as HPV-associated HNSCCs respond better to therapy. The RNAscope HPV-test is a novel RNA in situ hybridization (ISH) technique which strongly stains transcripts of E6 and E7 mRNA in formalin-fixed, paraffin-embedded tissue, with the potential to replace the indirect immunohistochemical (IHC) marker for p16 protein. A direct clinical comparison between p16 IHC and an automated RNA ISH using 18 probes has not been established. Samples from 27 formalin-fixed, paraffin-embedded HNSCC cases from the Emory University Hospital archives were stained using 18 individual RNA ISH probes for high-risk HPV (RNAscope 2.5 LS Probe ) on a Leica autostainer (Buffalo Grove, IL) and were compared with p16 IHC. Two pathologists reviewed and reached a consensus on all interpretations. The RNAscope technique was positive in 89% (24/27) and the p16 IHC was positive in 78% (21/27). The RNAscope was negative in 11.1% of samples (3/27) and the p16 IHC-negative in 22.2% (6/27). The RNA ISH detected 100% of the p16-positive IHC-stained slides and had a concordance of 88.9% (24/27). This easy to interpret automated staining method for 18 high-risk HPV genotypes is a feasible replacement for the indirect p16 IHC method.

Two Arabidopsis thaliana dihydrodipicolinate synthases, DHDPS1 and DHDPS2, are unequally redundant.

In Arabidopsis thalinana (L.) Heynh., DHDPS1 and DHDPS2 encode orthologous dihydrodipicolinate synthases (DHDPS), the first enzyme of the lysine (Lys) biosynthesis pathway. A TDNA insertion mutant of dhdps2 was previously reported to be viable and to accumulate free threonine (Thr). Analysis of additional TDNA insertion lines showed that dhdps1 and dhdps2 mutants are both viable and that whereas dhdps2 mutants accumulate Thr, dhdps1 plants do not. Thr-accumulation was complemented by heterologous expression of Escherichia coli DapA, indicating that the phenotype is due to reduced DHDPS activity in dhdps2. DHDPS1 contributes ~30% towards the total DHDPS activity in leaves of young plants and DHDPS2 contributes 70%; therefore, the threshold of activity resulting in Thr accumulation lies within this narrow range. dhdps1-dhdps2 double mutants could not be isolated, even after exogenous feeding with Lys. Segregation analysis indicated that gametes lacking functional DHDPS genes are defective, as are embryos. Plants carrying only a single DHDPS2 gene do not accumulate Thr, but they show a gametophytic defect that is partially rescued by Lys application. Despite the accumulation of Thr, dhdps2 seedlings are no more sensitive than wild-type plants to growth inhibition by Lys or the Lys precursor diaminopimelate. They also are not rescued by methionine at growth-inhibitory Lys concentrations. Exogenous application of Lys and methionine to dhdps2 mutants did not reduce the accumulation of Thr.