The contribution of CD200 to the diagnostic accuracy of Matutes score in the diagnosis of chronic lymphocytic leukemia in limited resources laboratories

The Use of CD200 in the Differential Diagnosis of B-Cell Lymphoproliferative Disorders

Background: B-Cell Lymphoproliferative Disorders (B-LPDs) are a group of heterogenous disorders characterised by the accumulation of B-cells in peripheral blood, bone marrow, lymph nodes and spleen. They have a variable disease course and outcome and many share similar features making differential diagnosis challenging. Therefore, accurate diagnosis is fundamental in particular for determining treatment options. Immunophenotyping by flow cytometry plays a crucial role in the diagnosis of B-LPDs. However, overlapping immunophenotyping patterns exist and the use of novel monoclonal antibodies has become increasingly important in immunophenotyping analysis. More recently differential expression of CD200 has been reported in various B-LPDs and that CD200 may improve the differentiation between chronic lymphocytic leukaemia (CLL) and mantle cell lymphoma (MCL). In this study CD200 expression is evaluated in different B-LPDs. Methods: A total of 100 samples were collected and analysed by immunophenotyping flow cytometry over a period of 1 year (2017-2018), by a panel of monoclonal antibodies including CD200. The percentage of CD200 and its expression intensity was evaluated and compared between different groups of B-LPDs. Results: All of the 50 cases of CLL expressed CD200 with moderate to bright intensity, 6 MCL cases lacked the expression of CD200. Furthermore, all 5 cases of hairy cell leukaemia (HCL) expressed CD200. Out of all B-LPDs evaluated, CD200 expression in HCL cases was noted to be the brightest. The other 39 cases were not found to be B-LPDs. Conclusion: CD200 has an important role in differentiating CLL from MCL, HCL has a consistent bright expression of CD200. By adding CD200 to the combinations of markers in routine testing panel, Immunophenotyping by flow cytometry can be an effective tool in the diagnosis of B-LPDs especially in cases with atypical immunophenotyping pattern. Our result support that CD200 can be added to routine testing panel as it is useful in differentiating them.

Use of a hybrid intelligence decision tree to identify mature B-cell neoplasms

BACKGROUND

Mature B-cell neoplasms are challenging to diagnose due to their heterogeneity and overlapping clinical and biological features. In this study, we present a new workflow strategy that leverages a large amount of flow cytometry data and an artificial intelligence approach to classify these neoplasms.

METHODS

By combining mathematical tools, such as classification algorithms and regression tree (CART) models, with biological expertise, we have developed a decision tree that accurately identifies mature B-cell neoplasms. This includes chronic lymphocytic leukemia (CLL), for which cytometry has been extensively used, as well as other non-CLL subtypes.

RESULTS

The decision tree is easy to use and proposes a diagnosis and classification of mature B-cell neoplasms to the users. It can identify the majority of CLL cases using just three markers: CD5, CD43, and CD200.

CONCLUSION

This approach has the potential to improve the accuracy and efficiency of mature B-cell neoplasm diagnosis.

Methodological and conceptual challenges to the flow cytometric classification of leukemic lymphoproliferative disorders

The diagnosis of leukemic B-cell lymphoproliferative disorders (B-LPDs) is made by integrating clinical, cytological, cytometric, cytogenetic, and molecular data. This leaves room for differences and inconsistencies between experts. In this study, we examine methodological and conceptual aspects of the flow cytometric classification of leukemic B-LPDs that could explain them. Among methodological aspects, we discuss (1) the different statistical tests used to select and evaluate markers, (2) how these markers are analyzed, (3) how scores are interpreted, (4) different degrees to which diagnostic information is used, and (5) and the impact of differences in study populations. Among conceptual aspects, we discuss (1) challenges to integrating different biological data points, (2) the under examination of the costs of misclassification (false positives and false negatives), and finally, (3) we delve into the impact of the lack of a true diagnostic gold standard and the indirect evidence suggesting poor reproducibility in the diagnosis of leukemic B-LPDs. We then outline current harmonization efforts and our personal approach. We conclude that numerous flow cytometry scores and diagnostic systems are now available; however, as long as the considerations discussed remain unaddressed, external reproducibility and interobserver agreement will not be achieved, and the field will not be able to move forward if a true gold standard is not found.

Challenges in translational machine learning

Machine learning (ML) algorithms are increasingly being used to help implement clinical decision support systems. In this new field, we define as "translational machine learning", joint efforts and strong communication between data scientists and clinicians help to span the gap between ML and its adoption in the clinic. These collaborations also improve interpretability and trust in translational ML methods and ultimately aim to result in generalizable and reproducible models. To help clinicians and bioinformaticians refine their translational ML pipelines, we review the steps from model building to the use of ML in the clinic. We discuss experimental setup, computational analysis, interpretability and reproducibility, and emphasize the challenges involved. We highly advise collaboration and data sharing between consortia and institutes to build multi-centric cohorts that facilitate ML methodologies that generalize across centers. In the end, we hope that this review provides a way to streamline translational ML and helps to tackle the challenges that come with it.

The potential differential diagnosis value and clinical significance of CD35 expression in B-chronic lymphoproliferative disorders

Background

Flow cytometry for immunophenotyping is the main method for diagnosing chronic lymphocytic leukemia (CLL). Differential diagnosis between CLL and other B-chronic lymphoproliferative disorders (B-CLPDs) is sometimes difficult. This study aimed to investigate whether cluster of differentiation 35 (CD35) could be a useful marker for the differential diagnosis of CLL and other B-CLPDs.

Methods

The CD35 expression on lymphoma cells from 516 B-CLPD patients (347 CLL, 169 other B-CLPDs) was investigated through flow cytometry analysis. Serum C3 and C4 levels in B-CLPD patients were also evaluated.

Results

The results showed that the expression percentage and mean fluorescence intensity of CD35 were reduced in CLL cases compared with other B-CLPD patients. Furthermore, CD35 <17% produced a sensitivity of 81.8% and a specificity of 88.4% for supporting the diagnosis of CLL. Additionally, the addition of CD35 to Matutes score improved the score's discriminative power. The sensitivity of the Matutes score was improved from 81.3% to 88.5%, and the accuracy was improved from 96.6% to 97.6%. Finally, 15.0% and 16.4% of CLL patients had defective serum C3 and C4 levels at diagnosis, respectively.

Conclusions

Evaluating CD35 expression could have potential differential diagnostic value in distinguishing CLL from other B-CLPDs, especially between CLL and mantle cell lymphoma (MCL).