Interobserver concordance of assessments of dysplasia and blast counts for the diagnosis of patients with cytopenia: From the Japanese central review study

Clinical impact and characteristics of erythroid dysplasia in adult aplastic anaemia: Results from a multicentre registry

Morphological dysplasia in haematopoietic cells, defined by a 10% threshold in each lineage, is one of the diagnostic criteria for myelodysplastic neoplasms. Dysplasia limited to the erythroid lineage has also been reported in some cases of aplastic anaemia (AA); however, its significance remains unclear. We herein examined the impact of erythroid dysplasia on immunosuppressive therapy responses and survival in AA patients. The present study included 100 eligible AA patients without ring sideroblasts. Among them, 32 had dysplasia in the erythroid lineage (AA with minimal dysplasia [mini-D]). No significant sex or age differences were observed between AA groups with and without erythroid dysplasia. In severe/very severe AA and non-severe AA patients, a response to anti-thymocyte globulin + ciclosporin within 12 months was observed in 80.0% and 60.0% of AA with mini-D and 42.9% and 90.0% of those without dysplasia, with no significant difference (p = 0.29 and p = 0.24 respectively). Overall survival and leukaemia-free survival did not significantly differ between the groups. Collectively, the present results indicate that the presence of erythroid dysplasia did not significantly affect clinical characteristics or outcomes in AA patients, suggesting that its presence in AA is acceptable. Therefore, erythroid dysplasia should not exclude an AA diagnosis.

Reducing cytogenetic testing in the era of next generation sequencing: Are we choosing wisely?

INTRODUCTION

In most laboratories, next generation sequencing (NGS) has been added without consideration for redundancy compared to conventional cytogenetics (CG). We tested a streamlined approach to genomic testing in patients with suspected myeloid and plasma cell neoplasms using next generation sequencing ("NGS first") as the primary testing modality and limiting cytogenetics (CG) to samples with morphologic abnormalities in the marrow aspirate.

METHODS

Based on morphologic interpretation of bone marrow aspirate and flow cytometry, samples were triaged into four groups: (a) Samples with dysplasia or excess blasts had both NGS and karyotyping; (b) Samples without excess blasts or dysplasia had NGS only; (c) Repeat samples with previous NGS and/or CG studies were not retested; (d) Samples for suspected myeloma with less than 5% plasma cell had CG testing cancelled.

RESULTS

Seven hundred eleven adult bone marrow (BM) samples met the study criteria. The NGS first algorithm eliminated CG testing in 229/303 (75.6%) of patients, primarily by reducing repeat testing. Potential cost avoided was approximately $124 000 per annum. Hematologists overruled the triage comment in only 11/303 (3.6%) cases requesting CG testing for a specific indication.

CONCLUSIONS

Utilizing NGS as the primary genomic testing modality NGS was feasible and well accepted, reducing over three quarters of all CG requests and improving the financial case for adoption of NGS. Key factors for the success of this study were collaboration of clinical and genomic diagnostic teams in developing the algorithm, rapid turnaround time for BM interpretation for triage, and communication between laboratories.

Assessment of dysplasia in bone marrow smear with convolutional neural network

In this study, we developed the world's first artificial intelligence (AI) system that assesses the dysplasia of blood cells on bone marrow smears and presents the result of AI prediction for one of the most representative dysplasia—decreased granules (DG). We photographed field images from the bone marrow smears from patients with myelodysplastic syndrome (MDS) or non-MDS diseases and cropped each cell using an originally developed cell detector. Two morphologists labelled each cell. The degree of dysplasia was evaluated on a four-point scale: 0–3 (e.g., neutrophil with severely decreased granules were labelled DG3). We then constructed the classifier from the dataset of labelled images. The detector and classifier were based on a deep neural network pre-trained with natural images. We obtained 1797 labelled images, and the morphologists determined 134 DGs (DG1: 46, DG2: 77, DG3: 11). Subsequently, we performed a five-fold cross-validation to evaluate the performance of the classifier. For DG1–3 labelled by morphologists, the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy were 91.0%, 97.7%, 76.3%, 99.3%, and 97.2%, respectively. When DG1 was excluded in the process, the sensitivity, specificity, PPV, NPV, and accuracy were 85.2%, 98.9%, 80.6%, and 99.2% and 98.2%, respectively.

Clinical value of next-generation sequencing compared to cytogenetics in patients with suspected myelodysplastic syndrome

Next-generation sequencing (NGS) increasingly influences diagnosis, prognosis and management of myelodysplastic syndrome (MDS). In addition to marrow morphology and flow cytometry, our institution performs cytogenetics (CG) and NGS-based testing routinely in patients with suspected MDS. We evaluated the relative value of NGS in the assessment of patients with suspected MDS. We initially compared the diagnostic and prognostic information derived from CG and NGS in 134 patients. NGS enhanced the diagnostic yield compared to CG for clonal myeloid disorders (sensitivity 77% vs. 42·2%; specificity 90·2% vs. 78%; positive predictive value 92·8% vs. 76%; and negative predictive value 70·8% vs. 45·5%). The identification of poor prognosis mutations by NGS altered risk category in 27/39 (69·2%) patients with MDS with good/intermediate risk CG. Subsequently, we prospectively evaluated 70 patients with suspected MDS using an 'NGS-first approach' with CG restricted to samples with morphological abnormalities. We rarely identified mutations or CG abnormalities in patients without dysplastic features. NGS has a superior diagnostic performance compared to CG in patients with suspected MDS. We estimate that by using an 'NGS-first approach' we could reduce karyotyping by approximately 30%.

Where Does Morphology Fit in Myelodysplastic Syndrome Diagnosis in the Era of Molecular Testing?

The recognition of cytologic dysplasia in blood and bone marrow remains the cornerstone of myelodysplastic syndromes (MDS) diagnosis because it distinguishes MDS from clonal hematopoiesis of indeterminate potential or clonal cytopenia of undetermined significance. Expert morphologists achieve high concordance in the diagnosis of MDS if appropriate clinical information is provided. Because of the low prevalence of MDS, diagnostic approaches based solely on molecular diagnosis will likely be erroneous.