Flow cytometry is widely used for B-cell precursor acute lymphoblastic leukemia (BCP-ALL) diagnosis, and the detection of Minimal Residual Disease (MRD) by flow cytometry is an established stratification tool after induction therapy. Early risk stratification in BCP-ALL allows for individualized treatment strategies, improving prognosis and lowering treatment related toxicity.

The Leukemia-Associated ImmunoPhenotype (LAIP) at time of diagnosis might comprise an early and easily accessible risk stratification tool. However, immunophenotypically distinct subpopulations are common in BCP-ALL (Øbro et al, Leukemia, 2011), yet current scoring methods do not account for this heterogeneity. This could make classification less accurate and hereby underestimate the prognostic value of the LAIP, urging the need for detailed LAIP characterization.


To systematically characterize the LAIP including immunophenotypic subpopulations at time of diagnosis in order to assess the prognostic potential of the LAIP in patients with BCP-ALL


We retrospectively evaluated flow cytometry data from time of diagnosis in 200 patients diagnosed with BCP-ALL between October 2009 and June 2015 and treated according to the NOPHO ALL 2008 protocol. Age, white blood cell count (WBC), cytogenetics and day 0 treatment stratification were registered.

The expression of intracellular and surface B-lineage markers (CD19, CD20, CD22, nTdT, cyCD79a, and cyCD22), non-lineage (CD45, CD34, CD38, CD10, and nTdT) and cross-lineage expressed markers (CD123, CD66c, CD133, CD13, CD33, and CD15) was scored as negative or positive (dim/normal/bright), and broad or bimodal expression was registered. For bimodal expression the overall score was assigned according to that of the dominant subpopulation. For reference intervals, five non-ALL bone marrow samples with unaffected B-lymphopoiesis were used. The CD34, CD38, nTdT and CD45 median fluorescence intensity was assessed (data will be presented). All samples were run on a FACS Canto and analyzed in DIVA 6.0 software.

End-of-induction (day 29) MRD by flow cytometry was used as primary outcome. The best combination of markers was found using a forward selection procedure in a multiple regression analysis. Gene chip analyses were performed in 171 of the 200 patients (human gene 1.0 ST array, Affymetrix) (data will be presented)


Immunophenotypically distinct subpopulations were common at time of diagnosis (54%) with 19 patients (9.5%) having three or more subpopulations. CD34 was the most commonly bimodal marker (bimodal in 32.7%).

We examined the association of the LAIP with MRD, accounting for the frequent bimodal marker expressions in LAIP scoring. Of all markers, CD34 expression had the strongest association with MRD (p<0.0001, linear regression. p=0.0034 adjusting for age, cytogenetic group, treatment arm and WBC).

A combination of CD34 (p=0.0005), CD38 (p=0.0288) and nTdT (p=0.0093) (multiple regression) predicted MRD best with a CD34pos/bright, CD38 neg/dim, nTdT neg/dim phenotype associating with higher MRD. In order to exclude marker expression differences related to cytogenetic groups, we further examined patients with no cytogenetic aberrations (n=73). The association of the three markers with MRD was confirmed in this group.

We examined whether high CD34 expression was enriched among patients who later relapsed (n=7), and this was the case: CD34 was brightly expressed in 42% vs. 5% in non-relapse patients, further suggesting an association between CD34 expression and prognosis.


By applying a detailed LAIP characterization accounting for subpopulations, we showed that malignant BCP blasts expressing an immature phenotype (CD34pos, CD38dim, nTdTdim), resembling the pre-pro B cell-like phenotype (Sanz et al, PNAS, 2010), seem to be associated with poor treatment response in BCP-ALL. The LAIP might be a useful tool for early treatment stratification in the future.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.

Sign in via your Institution