Expression Profiling of Pediatric Acute Lymphoblastic Leukemia (ALL) Blasts at Diagnosis Accurately Predicts both the Risk of Relapse and of Developing Therapy-Induced ALL
Presenter: E. Yeoh
PresentAer's Affiliation: National University of Singapore
Type of Session: Plenary
Pediatric patients with ALL (PALL) are typically stratified into risk groups to help direct therapy.
80% of patients are now cured of pediatric ALL with systemic chemotherapy and treatment of the CNS.
Despite this high cure rate, those patients at ultra low risk of relapse get therapy that is too aggressive, and often suffer neurologic problems and treatment related leukemia.
High risk patients, on the other hand, continue to do poorly. In the current study, investigators employ molecular techniques in attempt to better predict outcome in the PALL population.
Materials and Methods
Investigators utilized oligonucleotide microarrays to analyze the expression of 12000 genes in diagnostic bone marrow blasts from 393 PALL patients.
360 samples were successfully analyzed.
Using artificial neural networks, as well as other statistical techniques investigators were able to derive specific gene expression profiles for all of the known subtypes of PALL.
Based on this genetic information, investigators were able to separate patients into 2 groups with significantly different relapse-free survival (25% vs. 98%; p < 0.0001).
Gene expression profiles were also predictive of the development of therapy-induced AML.
These data suggest that expression profiling of leukemia blasts at diagnosis will significantly improve our ability to accurately direct therapy in children with ALL.
This study dramatically illustrates the power of molecular biology techniques that may improve clinical medicine.
Oncolink's ASH Coverage made possible by an unrestricted Educational Grant from Amgen.