Coagulation FXIII-A expression patterns can predict outcomes in pediatric BCP-ALL

Coagulation factor FXIII subunit A (FXIII-A) together with subunit B make up the plasma-circulating FXIII proenzyme. FXIII-A has been identified as a dimer expressed in multiple cell types and it seems to be involved in cytoskeletal remodelling, phagocytosis, chemotaxis, and cell adhesion among others.¹ Previous studies have reported the expression of FXIII-A exclusively by leukemic B-cell progenitor (BCP) lymphoblasts but not by mature normal or leukemic B cells, nor by normal BCPs.¹ Moreover, a small study in 55 pediatric patients with BCP-ALL reported that FXIII-A expression is associated with a survival benefit in children with BCP-acute lymphoblastic leukemia (ALL).

In a recent study by Bettina Kárai et al.¹ the correlation of three FXIII-A expression patterns to different BCP-ALL clinical outcomes were investigated in pediatric cohorts. The results were published in Cancers and are summarized below.

Study design

• Multicenter study that collected bone marrow samples from 408 children with BCP-ALL at transplantation centers in Hungary, Slovakia, Poland, and Austria between 2011–2018. Patients with Down syndrome or t(9;22) and infants were not included.
• Survival data were used only from patients receiving the same treatment in accordance with the ALL Intercontinental Berlin-Frankfurt-Münster (IC-BFM) 2009 protocol.
• Immunophenotyping was performed by flow cytometry (FC) and the following criteria were used for the definition of the three FXIII-A expression patterns:
  • FXIII-A-negative: BCP-ALL with < 20% FXIII-A⁺ lymphoblasts
  • FXIII-A-dim: BCP-ALL with 20–79% FXIII-A⁺ lymphoblasts
  • FXIII-A-bright: BCP-ALL with ≥ 80% FXIII-A⁺ lymphoblasts
• Genetic testing by fluorescence in situ hybridization (FISH) was performed in 301 patient samples and the following genetic risk classification was used:
  • Low-risk group: t(12;21), ETV6, RUNX1, or high hyperdiploidy (51–65 chromosomes)
  • Intermediate-risk group: t(1;19) or all other genetic subgroups
  • High-risk group: MLL translocations, intrachromosomal amplification of chromosome 21 (iAMP21), complex karyotype, near haploidy (23–29 chromosomes), or low hypodiploidy (< 45 chromosomes)
• ALL IC-BFM risk categories were defined mainly as per the 2002 protocol² with the addition of iAMP21 and low hypodiploidy as high-risk features (2009 protocol).
• FC-based measurable residual disease (MRD) risk classification by MRD load was as below:
  • Low risk: < 0.1%
  • Medium risk: between 0.1% and < 10%
  • High risk: ≥ 10%
• The ‘B-other’ genetic classification refers to BCP-ALL cases without the established recurrent genetic abnormalities but with other genetic alterations.

Key findings

• Of the total 408 patients, 137 had FXIII-A-negative, 189 had FXIII-A-dim, and 82 had FXIII-A-bright protein expression.
• While there was a significant event-free survival (EFS) benefit for patients with FXIII-A-dim expression (p = 0.001), EFS rates were similar between the FXIII-A-negative and FXIII-A-bright patients. Also, overall survival (OS) was significantly improved for the FXIII-A-dim group, while it was similar for both the FXII-A-negative and -bright groups. The 5-year EFS and OS outcomes for each group are shown below in Table 1.

Table 1. Outcomes by FXIII-A expression level¹

• Univariate analysis for the FXIII-A-negative vs the FXIII-A-dim groups revealed that the following factors were significantly influencing EFS and OS (Table 2):
  • Age, prednisone response, ALL IC-BFM 2009 risk classification (high vs standard), genetic risk categories, and the presence of ‘B-other’ genetic abnormalities.
• In the multivariate analysis of the same groups, only genetic risk categories were a significant effector of EFS (HR, 5.78; 95% CI, 1.63–20.52; p = 0.007 good vs intermediate; HR, 17.4; 95% CI, 2.88–104.8; p = 0.002 good vs high) and OS (p = 0.031 good vs intermediate; p = 0.006 good vs high).
• When comparing the FXIII-A-bright vs the FXIII-A-dim groups significant EFS and OS influencers according to the univariate analysis were (Table 2):
  • FXIII-A expression pattern, ALL IC-BFM 2009 risk classification (high vs intermediate), and genetic risk categories.
• Multivariate analysis of the FXIII-A-bright vs the FXIII-A-dim groups revealed that FXIII-A expression patterns (p = 0.009) were significantly influencing EFS (with a trend for OS, p = 0.051), while genetic risk categories influenced only OS (p = 0.032 good vs intermediate; p = 0.031 good vs high).
• When comparing the FXIII-A-negative vs FXIII-A-bright groups, poor prednisone response was more common in FXIII-A-negative patients while the incidence of genetically high-risk patients was higher in the FXIII-A-bright group.

Table 2. Univariate analyses between FXIII-A expression groups for EFS and OS outcomes¹

• Pearson’s chi-square test and multinomial logistical regression analyses did not reveal any correlations between the different XFIII-A expression patterns and the risk stratification groups based on either the ALL IC-BMF 2009 or FC-based MRD classifications.
• Nevertheless, the intermediate genetic risk category and the iAMP21 abnormality were represented significantly more in the FXIII-A-negative group than in the -bright (p = 0.039 and p = 0.029, respectively) or -dim group (p = 0.009 and p = 0.029, respectively).
• Moreover, patients with FXIII-A-dim (OR = 0.49) or -bright (OR = 0.36) expression pattern had significantly lower chance of having a ‘B-other’ genetic alteration.

Conclusion

The pediatric patient subpopulation with dim FXIII-A protein expression has significantly longer EFS when compared to those with bright or negative FXIII-A expression. Moreover, the FXIII-A-dim group was associated with better OS when compared to patients negative for FXIII-A expression. This was linked by the authors to the statistically higher prevalence of patients with intermediate genetic risk and with ‘B-other’ genetic abnormalities in the FXIII-A-negative group. The results of this study indicate that FXIII-A negativity is a potential prognostic marker for survival outcomes and might be used as an additional risk stratification marker in upcoming trials.