Ras Pathway Activation in Childhood B-Lineage Acute Lymphoblastic Leukemia: Ras, Ptpn11, and Flt3 Mutations.

Background: The relatively low incidence of Ras mutations in acute lymphoblastic leukemia (ALL) has suggested that Ras pathway plays only a minor role in the disease. However, the recent finding of activating mutations in PTPN11, which encodes SHP-2 phosphatase, FLT3, or suggest that Ras can also be activated through upstream mechanisms in childhood ALL. Assessing the frequency, spectrum, and relation with clinical features of these mutations should permit to better evaluate the implication of Ras pathway activation.

Methods: 209 children (aged 1 to 17 yrs) with B-lineage ALL were studied at diagnosis. ALL with a t(12;21) were excluded from the study since no mutation of the Ras pathway has been found previously in this type of ALL. All patients were enrolled in the EORTC 58951 trial and the median follow up was 3 years. Relapse samples were also studied in 20/26 children who relapsed. Mutations of PTPN11, N-Ras, and H-Ras were screened by direct sequencing. FLT3 length mutations were screened by capillary electrophoresis, and FLT3 activating loop mutations (D835/I836) by EcoRV digestion of PCR products. FLT3 mutations were further caracterized by sequencing.

Results:N-Ras or K-Ras was mutated in 28 patients (13%). PTPN11 mutations were found in 10 (5%) cases (9 in exon 3, 1 in exon 13) and were absent at remission, excluding a Noonan syndrome in these patients. FLT3 was mutated in 6 (3%) cases (one internal tandem duplication, and 5 ins/del). Other specific mutations were MLL rearrangement in 2 (1%) cases, E2A-PBX1 in 10 (5%) cases, and BCR-ABL in 8 (4%) cases. With exception of one patient having both E2A-PBX1 and a N1-Ras mutation, genetic deffects were never found together. No difference was observed at presentation between ALL with Ras, PTPN11 and FLT3 mutations and other ALL regarding age, gender, WBC, or risk group. A trend toward over-representation of hyperdiploid caryotype was observed in mutated ALL (58% vs 46% in non mutated ALL) but it was not significant. No difference was either observed in response to prephase, minimal residual disease or relapse rate. The Ras mutation found at presentation was lost in 3 out of 4 patients who relapsed. In one of these patients, a new Ras mutation appeared at relapse. In an additional patient Ras mutation was only found at relapse.

Conclusion: We confirmed the presence of FLT3, and PTPN11 activating mutations in childhood ALL. However, mutation of either of these genes was only found in 21% of TEL-AML1 negative B-lineage childhood ALL, which is lower than previously described. The mutual exclusion of t(9;22), Ras, PTPN11, and FLT3 mutations can be explained by their partially redundant effect on Ras pathway and is in favor of their implication in leukemogenesis. The frequency of mutations activating directly or undirectly Ras pathway suggests that, contrary to what what was previously thought, Ras could represent a major pathway in ALL Surprisingly, while FLT3 and PTPN11mutations remained stable during the course of the disease, this was not the case Ras for mutations. This could explain the lack of prognostic relevance of these mutations and possibly reduce the interest of treatment strategies specifically targeting Ras in ALL.