The Use of Monocyte Subset Repartitioning By Flow Cytometry for Diagnosis of Chronic Myelomonocytic Leukemia

Introduction Patients with chronic myelo-monocytic leukaemia (CMML) have been reported to have a relative predominance of classical or MO1 monocytes (CD14+/CD16-) at the expense of MO2 (CD14low/CD16+) and MO3 (CD14-/CD16+) monocytes (Selimoglu-Buet. Blood 2015). These authors suggested that an MO1 percentage cut-off of >94% could predict the diagnosis of CMML with high sensitivity and specificity (both >90%) from other causes of monocytosis. Since then, several independent groups have attempted to reproduce their protocol with variable results. Most recently, the Mayo Clinic (Pophali. Blood Cancer J. 2019) found that an MO1 cut off of > 94% in peripheral blood identified CMML with a sensitivity of 75% and a specificity of 95.4%. These figures were lower than previously reported - calling into question, the utility of flow cytometry to distinguish the aetiology of monocytosis in a real-world setting. Objective Our retrospective audit aimed to establish if monocyte subset repartitioning could be used to reliably diagnose CMML in a real-world sample of patients with monocytosis. Methods In this study, we assessed peripheral blood samples from 35 patients presenting with a monocytosis (absolute monocyte count > 1 x109/L) in a tertiary referral hospital in Brisbane, Australia between June 2015 and Sep 2019. The patients' final clinical diagnosis was extracted from the medical record by two clinicians (AM, PM) blinded to the results of the flow cytometry. Peripheral blood samples were subjected to MFC at a median of <24 hours (range, < 24 hours to 160 hours) after collection. Flow cytometry was performed using the BD FACS Canto II flow cytometer. Monocyte subsets were identified using Kaluza software (Beckman Coulter, USA). A CD45/ side scatter gate was set to locate the monocyte population and specific antibody combinations were used to identify and exclude other lineages; these were - CD24 to exclude granulocytes and B cells, CD16 to exclude neutrophils, CD2 to exclude T cells and CD56 to exclude NK cells. Based on the CD14 and CD16 expression, the monocytes were then divided into: MO1 (CD14+/CD16-), MO2 (CD14low/CD16+) and MO3 (CD14-/CD16+). Results are reported descriptively and Fishers Exact Test (IBM® SPSS® Statistics Version 26) was used to establish if there was any correlation between the percentage of classical monocytes and the diagnosis of CMML. Results Of the 35 patients included, 13 patients had CMML and four patients were diagnosed with another underlying myeloid neoplasm: one patient with myelodysplastic syndrome (MDS-RCMD), one patient with myeloproliferative neoplasm - not otherwise specified (MPN-NOS), one patient with multiple myeloma and one patient with acute myeloid leukaemia with myelodysplasia related changes (AML-MRC). Eighteen cases had non-clonal monocytosis. Six patients had a reactive monocytosis in the setting of autoimmune disease including: granulomatosis with polyangiitis (n=2), rheumatoid arthritis (n=1), IgG4 disease (n=1), mixed connective tissue disease (n=1) and polyarticular gout (n=1). Twelve cases of reactive monocytosis were due to infective and inflammatory causes. Among our 13 patients with CMML, seven cases (53.85%) had an MO1 percentage > 94%. In comparison, only four (18.18%) among the 22 cases of non-CMML were identified to have an MO1 percentage > 94%. There was no correlation between an MO1 percentage cut off of > 94% and a diagnosis of CMML (p = 0.057). Our study also examined the utility of an MO3 percentage cut off of < 1.13%, established by Hudson et al (Am J Clin Pathol 2018). Among our 13 cases of CMML, six patients (46.15%) were noted to have an MO3 percentage < 1.13%. Meanwhile, only eight, out of 22 non-CMML patients (36.36%) had MO3 percentage < 1.13%. There was no association between MO3 percentage and the diagnosis of CMML (p = 0.724). Conclusion Using the MO1 and MO3 percentage cut offs previously established, we were unable to reliably diagnose CMML. Given these findings, we suggest that more research with larger sample sizes is required before monocyte subset analysis can be applied in the clinical laboratory to discriminate reactive monocytosis from CMML. Disclosures Mollee: Janssen: Membership on an entity's Board of Directors or advisory committees, Research Funding; BMS/Celgene: Membership on an entity's Board of Directors or advisory committees; Amgen: Membership on an entity's Board of Directors or advisory committees; Takeda: Membership on an entity's Board of Directors or advisory committees; Pfizer: Membership on an entity's Board of Directors or advisory committees; Caelum: Membership on an entity's Board of Directors or advisory committees.