Page 39 - 2021_06-Haematologica-web

Monocytic switch in pediatric BCP-ALL subtypes
were unable to uncover the genetic background. In our pre- vious work, we observed a slower response to initial treat- ment in patients with a monocytic switch than in patients without it. Despite significant changes in the phenotype towards the monocytic lineage in some of the patients, risk- based ALL treatment remained the treatment of choice.1
Recently, new genetic subtypes of ALL, particularly with- in the “B-other” group, were discovered using RNA sequencing (RNA-seq).2,3 We investigated whether any of the newly defined subsets had a higher tendency to under- go monocytic switch.
Monocytic switch is accompanied by the gradual loss of CD19. As the determination of minimal residual disease (MRD) by flow cytometry (FC) relies on B-cell markers, switching leads to MRD underestimation. The FC MRD value on day 15 is used for patient stratification in the cur- rent pediatric Berlin-Frankfurt-Münster (BFM) BCP-ALL treatment protocols. It is an open question whether FC monitoring of MRD should be adapted for patients with switch.
Recently, anti-CD19 therapy, namely, blinatumomab, was added to pediatric ALL frontline treatment protocols, and a larger proportion of patients will be treated with it in the near future. As patients with switch that includes CD19 loss may be selected for anti-CD19 treatment, knowledge about subtypes prone to switch will be of interest. Moreover, anti-CD19 therapy may result in monocytic switch, as was repeatedly reported, especially for cases of KMT2A rearrangement.4–9
The aim of this study is to describe the molecular land- scape of ALL with monocytic switch in the context of newly discovered genetic subtypes. We aim to describe the switching behavior in distinct genetic subtypes and to ana- lyze the extent of its influence on the FC assessment of MRD. In addition, we analyzed the impact of switching phenomenon on relapse risk.
Methods
This study was approved by the Institutional Review board of the University Hospital Motol and the Second Faculty of Medicine and informed consent was obtained from all patients and their guardians in accordance with the Declaration of Helsinki.
We included 726 BCP-ALL patients (age, 0-18 years) diagnosed in the Czech Republic between 09/2007 and 02/2019. Patients were treated according to the following BFM protocols: ALL-IC- BFM-2002 (n=17); ALL BFM 2000 (n=177); ALL BFM 2009 (n=483); the Interfant 2006 protocol (for children younger than 12 months; n=30); the EsPhALL protocol (for BCR-ABL1-positive patients; n=16); and other protocols (n=3; one patient was treated by a modified protocol, one Down syndrome patient with signif- icant comorbidities was treated with a reduced ALL BFM 2009 protocol, and one patient moved abroad after induction therapy). Some of the 726 patients (179 of 726) diagnosed between 09/2007 and 05/2010 were included in a previous study.1 For selected analyses, we expanded this consecutive cohort with additional 19 patients who were diagnosed before 09/2007 and for whom the switching phenomenon was identified and described retrospec- tively (n=16)1 or who were diagnosed abroad (Germany, n=2; and Slovakia, n=1) (Online Supplementary Figure S1). All patients were screened for the presence of recurrent fusion genes (BCR-ABL1, KMT2A-AFF1, ETV6-RUNX1 and TCF3-PBX1). In all patients, standard cytogenetic evaluation and assessment of the DNA index were performed as published previously.1,10 The B-other subset
was defined as BCP-ALL by the absence of all routinely investigat- ed classifying aberrations (ETV6-RUNX1, hyperdiploidy, hypodiploidy, BCR-ABL1, KMT2A rearrangements and TCF3- PBX1).
Flow cytometry
The diagnostic phenotype was determined through standard protocols.11–13 Ambiguous lineage acute leukemia (ALAL) diagnosis was based on the European Group for the Immunological Characterization of Leukemias (EGIL) criteria11,12,14,15 and/or the World Health Organization16 classification. A summary of the antibody clones and vendors is presented in the Online Supplementary Table S1. In addition, the percentage of the B-mono- cytoid population defined as CD19posCD14pos was determined with an eight-color combination of antibodies against CD45, CD14, CD34, CD19, CD33, CD20, CD10 and CD3 at diagnosis (day zero [d0]), day 8 [d+8], day 15 [d+15], and day 33 [d+33] in the bone marrow (BM) and/or peripheral blood (PB) as shown pre- viously.1 FC-assessed MRD was evaluated using three- or four- color monoclonal antibody combinations in the period between 1998 and 200717 and using eight-color combinations starting in 2007.18,19 The sensitivity (level of quantification and level of detec- tion, LoQ and LoD, respectively) of the FC-assessed MRD was defined by the number of nucleated cells measured in an MRD- specific tube (for a sensitivity of 10-3 and 10-4, 20,000 and 200,000 nucleated cells were measured, respectively). Only samples with an appropriate FC MRD sensitivity were included in the analysis: generally, a sensitivity of 10-4 was required; for cases with poly- merase chain reaction (PCR)-determined MRD ≥10-2, measures of sensitivity that were at least one log value lower than that of the actual PCR-determined MRD value were sufficient. The sample was assessed as FC MRD positive when a cluster of at least 20 events with an aberrant B-cell phenotype was detected.
Definition of the switching phenomenon
Based on our previous work,1 we defined the switching phe- nomenon as the presence of an intermediate B-monocytoid popu- lation, i.e., BCP-ALL blasts with a gradual decrease in CD19 expression accompanied by a gradual increase in the expression of at least one monocytic marker (CD14, CD33, or higher side scat- ter, SSC) (Online Supplementary Figure S2) at any time point between d0 and d+33. We used fluorescence-activated cell sorting (FACS) of the intermediate B-monocytoid (CD19posCD14pos) and monocytoid (CD19negCD14pos) populations to show that the IG/TR rearrangements were identical to those in B lymphoblasts when enough material was available (patients n =37; samples n=70) between d0 and d+33.
Immunoglobulin/T-cell receptor polymerase chain reaction-assessed minimal residual disease
Patient-specific IG/TR assays were developed as described pre- viously.1,20,21 In patients with two independent IG/TR targets, the higher value was reported. Sensitivity (LoD) and quantitative range (LoQ) for each assay was defined according to the European Study Group on MRD guidelines.22 A minimum LoD of 10-4 was achieved in all patients/timepoints.
Statistics
Fisher’s exact test was used for comparing categorical variables, and the Mann-Whitney test was used for continuous variables. Other tests used are explicitly indicated in the text. The results were considered significant when P-values were less than 0.05. Statistical analyses were performed using GraphPad software (GraphPad Software, Inc., La Jolla, CA, USA), R23 and StatView version 5.0 (StatView Software, Cary, NC, USA).
haematologica | 2021; 106(8)
2067