Introduction: Subunit A of blood coagulation factor XIII (FXIIIA) may be expressed in leukemic B-cell precursor (BCP) lymphoblasts in addition to platelets, megakaryocytes, monocytes and macrophages (Kiss F. et al Thromb Haemost 2006;96:172-82.). In a retrospective single-center cohort of children with acute lymphoblastic leukemia (ALL) treated with the BFM ALL-IC 2002 protocol, expression of FXIIIA was correlated with statistically significant survival advantage (Kárai B et al. Pathol Oncol Res 2018; 24:345-52.). Our aim was to investigate the impact of FXIIIA expression pattern on EFS and its correlation with known clinical and genetic prognostic factors in a multi-center pilot study within the frames of the BFM ALL-IC 2009 clinical trial (EuDract: 2010-019722-13).

Patients and Methods: We examined 317 children with BCP-ALL between 2011 and 2018 at institutions in Poland (n=188), Hungary (n=116), and Slovakia (n=13). Patients with Down-syndrome, t(9;22), and infants (<1 year) were excluded. Immunophenotype and minimal residual disease (MRD) were determined by flow cytometry (FC). Cytogenetic analysis and fluorescence in situ hybridization were performed according to standard methods. DNA extracted from bone marrow samples was processed for SALSA multiplex ligation-dependent probe amplification P335-B2 ALL-IKZF1 probe mix (MRC-Holland, Amsterdam, The Netherlands) analysis according to manufacturer's instructions. Statistical analysis: Survival analysis was carried out by the Kaplan-Meier estimator. More than 2 groups were analyzed by the Kruskal-Wallis test. Dunn's multiple comparison was applied as post host test. Dichotomous categorical variables were compared by Chi square test and logistic regression to analyze multiple variables. P<0.05 was considered significant.

Results: We observed 3 different patterns of FXIIIA expression in leukemic lymphoblasts: negative pattern (<20%), moderate positive expression (20-79%), and strong positive pattern (>80%). Evaluation of FC histograms of the moderate positive pattern showed that FXIIIA expression increased continuously which excluded the existence of distinct FXIIIA negative and FXIIIA positive subpopulations within the moderate positive expression group (Fig.1). In contrast to the previous single-center retrospective cohort, EFS of patients with FXIIIA positive BCP-ALL was not significantly different from that of patients of FXIIIA negative BCP-ALL (Fig.2A). However, a significant EFS advantage of patients with moderate FXIIIA positivity was demonstrated when compared with patients with FXIIIA negative BCP-ALL (p=0.019), and with patients with FXIIIA strong positive BCP-ALL (p=0.001) (Fig.2B). The 3 different FXIIIA expression patterns did not correlate significantly with either risk stratification according to BFM ALL-IC 2009 or FC MRD categories. Intermediate genetic risk categories (low hyperdiploidy, t(1;19), and "B-other") were significantly more prevalent (p=0.042) among patients with FXIIIA negative BCP-ALL than among patients with strong FXIIIA positive BCP-ALL. The "B-other" subgroup was also significantly more prevalent (p=0.022) among patients of the above 2 groups (Table 1). Distribution of the "B-other" genetic group and the prednisone response differed significantly between the FXIIIA positive and negative groups. Intermediate genetic risk group is mostly made up of patients assigned to the "B-other" group explaining the significant difference between the FXIIIA positive and negative groups, in terms that FXIIIA negative patients had 50% lower chance of facing low instead of intermediate genetic risk compared to FXIIIA positive patients. The multivariate logistic regression analysis confirmed the association between the FXIIIA characteristics and the intermediate genetic risk group. This association persisted after adjusting for categorical variables, i.e. gender, age, WBC, and BFM ALL-IC 2009 stratification (Table 2). High risk copy number alterations were detected in 4/7 samples of the FXIIIA negative group and in 1/17 of the FXIIIA positive group.

Conclusions: FXIIIA expression status of leukemic lymphoblasts can easily be determined by FC. Patients with FXIIIA negative lymphoblasts should be further investigated with sophisticated and more expensive genetic and molecular methods.

Grant sponsor: OTKA K108885.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.

Sign in via your Institution