Newly identified genes may help predict outcome in childhoo...( al Children's Oncology Group and the Na...)

Current treatments, usually combination chemotherapy and post-induction therapeutic intensification (increasingly stronger treatment administered after initial therapy), currently help 75 percent of children with ALL achieve long-term remission. Yet, 25 percent of patients relapse with resistant disease. Additionally, 25 percent of patients who receive dose intensification treatment are often over-treated and may be cured using less intensive regimens with fewer acute and long-term side effects.

To identify strongly predictive genes, researchers performed gene expression profiling in two large, statistically designed, retrospective groups of pediatric ALL patients registered to trials previously conducted by the Pediatric Oncology Group (now merged with other national groups into the Children's Oncology Group) a group of 127 infant leukemia patients and a case control study of 254 children with B-precursor and T-cell ALL. Researchers used unsupervised learning tools and supervised machine learning algorithms to identify novel genes that were predictive of outcomes. Three strong genes were identified with both testing methods G0, an expressed sequence tag of previously unknown function; G1:GNB2L1, a G-protein (a second messenger receptor of intracellular response) and activator of protein kinase C, which plays a pivotal role in cell signaling systems; and G2, an interleukin (IL)-10 receptor alpha, which regulates immune and inflammatory response. ALL cases expressing higher levels of these genes were associated with better outcomes. As the group fully cloned and characterized the G0 expressed sequence tag, it was given the new name of OPAL1.

"Studies to profile gene expressions in cancers like leukemia are very timely be
'"/>

Contact: Aimee Frank
amf@spectrumscience.com
202-955-6222
American Society of Hematology
7-Dec-2003