Abstract

ProB ALL is considered an unfavorable subset of ALL both in children and adults. It is currently defined as B precursor ALL with no expression of CD10. MLL/AF4pos cases, who are almost never express CD10, contribute to the poorer prognosis of proB ALL. Until now, it has not been solved whether the prognosis of proB children remains poorer even after removal of MLL/AF4pos cases. We have analyzed the prognosis of proB ALL in comparison to the other types of B precursor ALL in a population-wide cohort of Czech children tested in our lab between 9/1996 and 8/2006; we asked whether the prognosis was affected by MLL/AF4pos cases. Concurrently, we asked whether a more sensitive definition of proB ALL uncovers additional patients with a similar prognosis. We retrospectively analyzed all 505 patients tested in our reference lab, age below 18 years at diagnosis of B precursor ALL. All patients have been treated in one of the Czech Pediatric Hematology working group centers according to the standard treatment protocols BFM95 (n=290), Interfant99 (n=12), Interfant2006 (n=3), ALL-IC BFM 2002 (n=192) or POG9407 (n=5); 3 children were treated/followed off-protocol. ProB ALL (n=42) was associated with a poorer prognosis in the entire cohort (p= 0.0084, 5y EFS 62%±7.9%). After removing MLL/AF4pos ALL (n=10), 5y EFS is 66% ± 8.8% and the difference from CD10pos patients is no longer significant (p=0.09). Next, we tested a hypothesis that patients with a “partial proB” ALL (i.e., those containing CD10pos cells in addition to a substantial proportion of CD10neg lymphoblasts) have a poorer prognosis. The percentage of CD10neg cells was defined in all as a difference between CD19pos and CD10pos gated lymphoblasts; cases with greater than 20% of CD10neg cells who did not fulfill the current proB ALL definition were categorized as “partial proB”. The prognosis of “partial proB” ALL (n=45; 8.9%) is similar (5y EFS 63% ± 7.7%) to the classical proB ALL. With the exception of MLL/AF4, none of which expresses CD10, the distribution of main genotype categories is not significantly different between proB ALL and “partial proB” ALL. In contrast, the genotype distribution differs between “partial proB” ALL and other non-proB cases (p=0.0072). Response of the “partial proB” patients to prednisone is not different from the “classical” proB but poorer than that of non-proB ALL (pFisher=0.0039). Combining the “partial proB” and proB ALL cases into one category (n=87) defines a subset with a poor prognosis (n=77, 5y EFS 64%±5.9%), which is significantly poorer than that of other B precursor ALL cases even after excluding MLL/AF4pos children (5y EFS 78%±2.2%; p=0.011). Therefore, the definition of proB ALL should be based on the percentage of CD10neg blasts, rather than on a complete absence of CD10 in all blasts–such criteria define patients with a new ALL subtype whose prognosis is poorer even after excluding MLL/AF4pos ALL.

Supported by MSM0021620813, MZdNR/9531–3. Czech Pediatric Hematology Working Group.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author