Background. ALL is the most frequent cancer in childhood. Although current chemotherapy protocols reach an event free survival (EFS) greater than 70%, the remaining cases relapse and ALL is still the first cause of death in children with cancer. Several efforts have been made to better understand the underlying mechanism involved in ALL. Current chemotherapy protocols have reached a plateau of effectiveness. The detection of minimal residual disease (MRD) during the course of treatment will definitively allows us to stratify children with ALL in standard (SR), intermediate (MR) and high (HR) risk, respectively. In the latter group we included patients with a prednisone poor response, with t(9;22) and t(4;11) (HR by default). Despite of this better definition of therapeutic response, the majority of relapses are included in the MR group, suggesting that we are treating different forms of ALL in this subset. For this reason new biomarkers and molecular targets are highly mandatory. The objective of functional proteome analyses of ALL is to bridge clinical applications with new biomarker discoveries that will be useful for determining prognosis and response to therapy and to identify potential new more selective treatments.

Materials and Methods. We analyzed 81 diagnostic samples (peripheral blood and/or bone marrow) with B precursor ALL, that were diagnosed and treated at our centers (Catania and Padova) and enrolled in the AIEOP-LLA 2000 protocol. Reverse phase proteomic assay (RPPA) was used to interrogate the expression of 81 phosphorylated or native protein endpoints in our patients’ samples. We evaluated Pro- and Anti-Apoptotic, Protein Kinase, and Growth factor receptor cell signaling pathways. For statistical purposes, we considered the following characteristics: age at diagnosis, gender, immunophenotype, karyotype, white blood cell count and percentage of blast at diagnosis, response to prednisone, level of MRD at day 33 (TP1) and day 78 (TP2) during the induction phase. We distributed patients in 4 groups, based on risk (SR 21, MR 31, HR 10) and occurrence of relapse (19). We also studied patient specimens both at diagnosis and relapse only in those samples with a blast rate greater than 50% (10 cases out of 19). All findings obtained with RPPA were confirmed by Western blot analyses. We applied the Wilcoxon Test and the t-Test, using the ‘R’ software. Both univariate and multivariate analyses were performed.

Results. Comparing subgroups for gender, WBC count (higher vs less than 50.000/ml), immunophenotype (Common vs others), karyotype (t(12;21) positive vs negative), prednisone good (PGR) vs poor (PPR) responders, analysis of MRD (SR vs MR, MR vs HR, SR vs HR, MR vs MR relapsed) and diagnoses vs relapses, we did not find any statistically significant data. Conversely, we found a statistically significant higher expression of FosB and Annexin II in the group of children with WBC >50.000/ml (p value <0,05); and of Zap70 (Y319)/Syk (Y352) in the subgroup of MR cases who did not suffer from relapse (p value <0,05). Our study showed that in the apoptotic pathway, we detected a higher expression of all the pro-apoptotic proteins [Bax, SMAC/DIABLO, Caspase 7 total and cleaved (D198)], a low expression of anti-apoptotic proteins (Bcl-xL, XIAP) associated with a reduction or a lack of PARP total or cleaved (D214). We also found that a specific pattern of expression (higher PTEN (S380), PARP cleaved, Caspase 7 cleaved, PDK (S241), PKAc (T197), p90RSK (S380), MEK ½ (S217–221), IKBA (S32), GRB2; lower Beta Catenine) selected a subgroup of children (n°9) with a better survival when compared with other cases (p value <0,05).

Conclusions. Our preliminary data showed that RPPA in association with Western blot analysis, is a reliable strategy for the identification of biomarkers and/or molecules for targeted therapy. Application of this methodology on specific subgroups of children with ALL will allow us to identify new therapeutic strategies.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author

Sign in via your Institution