Assessment of Dysgranulopoiesis By Optical Cytology Compared to Distribution of Monocyte Subsets By Flow Cytometry in Peripheral Blood for Predicting the Diagnosis of CMML

Introduction

The diagnosis of chronic myelomonocytic leukemia (CMML) according to the WHO 2017 requires the presence of absolute (≥ 1x109/L) and relative (≥ 10%) monocytosis. Optical cytology represents the basis for its diagnosis and is essential to evaluate dysmyelopoiesis and to identify promonocytes and blasts. Although dysplasia is not a specific finding, its presence along with a persistent monocytosis is highly suggestive of CMML. In this sense, dysgranulopoiesis is an almost constant finding in CMML. The study of monocyte subsets distribution in peripheral blood (PB) by multiparametric flow cytometry (MFC) has proven to be a very useful tool, since the presence of >94% of classical phenotype monocytes (MO1) allows to predict the diagnosis of CMML with a sensitivity and specificity greater than 90% (Selimoglu-Buet et al. Blood 2015). The aim of this study was to evaluate the predictive capacity of morphological evaluation of PB smears for the diagnosis of CMML in a series of patients with absolute and relative monocytosis. Additionally, these results were compared with the analysis of monocyte subsets distribution in PB by MFC.

Methods

162 patients with absolute and relative monocytosis were studied. 114 PB smears were available for morphological assessment and 160 MFC analysis of monocyte subsets distribution. The final diagnosis for these patients was: CMML (n=76), Phi negative chronic myeloproliferative neoplasia (MPN) (n=9) and reactive monocytosis (n=77). Genomic information (cytogenetics and NGS) was available in 66 CMML and all of MPN. After anonymizing PB smears, a cytological assessment was performed without additional clinical or analytical data (eg.: age, blood counts, biochemistry). This way we tried to compare the predictive capacity for the diagnosis of CMML of both techniques without the influence of additional data other that the known absolute and relative monocytosis. The cytological evaluation of PB smears consisted fundamentally of a qualitative evaluation of dysgranulopoiesis: nuclear hyposegmentation, chromatin hypercondensation and/or cytoplasmic hypo-degranulation. The sensitivity, specificity, Youden index, positive predictive value (PPV) and negative predictive value (NPV) of both techniques were analyzed. The Youden index allows to evaluate the balance between sensitivity and specificity ([sensitivity + specificity] -100). The capacity for diagnostic prediction of CMML for both techniques was assessed using the C-index. Finally, Cohen's Kappa index was used to evaluate the concordance between both techniques.

Results

Table 1 shows the results of sensitivity, specificity, Youden index, PPV, NPV and C-index of both techniques. The Kappa index showed a very good degree of agreement (K=0.715; p<0.001) between both techniques.

Conclusions

The presence of >94% MO1 had a better balance between sensitivity and specificity than the assessment of dysgranulopoiesis in PB. It also had a high sensitivity, demonstrating that it can be very useful screening tool. Despite this, the specificity and PPV of the assessment of dysgranulopoiesis were higher than those of the MFC, demonstrating that the identification of morphological dysplasia would allow the diagnosis of CMML to be predicted with a higher degree of certainty in the initial workup of absolute and relative monocytosis.

No relevant conflicts of interest to declare.