Identification of risk markers in Acute Lymphoblastic Leukemia.

D Khurana BS^a, J Qiu MS MD^b, P Gunaratne PhD^a, JF Margolin MD^b, MC Gingras PhD^b

^aDepartment of Molecular and Human Genetics, Baylor College of Medicine, Human Genome Sequencing Center, Houston, Texas, United States, ^bDepartment of Pediatrics, Baylor College of Medicine, Texas Children’s Cancer Center, Houston, Texas, United States

BACKGROUND: There are large differences in outcomes between adult and pediatric patients with Acute Lymphoblastic Leukemia (ALL), as well as between pediatric ALL patients with different chromosomal translocation. These differences are obvious among clinical risk groups. In the lowest risk groups, pediatric patients are cured at a rate of 80-90% but only 50% of adult patients are cured. In higher risk categories, approximately 60% of children but only 10% of adults are cured. We hypothesize that the genes differentially expressed in adult versus pediatric and in lower- versus higher-risk pediatric ALL are involved in mechanisms of drug resistance or drug responsiveness. AIM: Identify risk-associated genes that might control response to treatment. METHODS: We used an integrated approach to facilitate gene expression profiling that combines suppression subtractive hybridization (SSH), concatenated cDNA sequencing (CCS), and quantitative real-time RT-PCR (QR-RT-PCR). We have generated specific high-risk and low-risk ALL gene expression profiles by subtracting pools of cDNA from adult and pediatric t(9;22) ALL against each other. To investigate the risk associated value of these genes, we have measured the expression level of these genes the diagnostic samples for a cohort of 51 ALL patients. The patient distribution included 18 low-risk ALL patients [hyperdiploid and t(12;21)] and 33 high-risk ALL [adult and pediatric t(9;22), pediatric t(1;19) and t(4;11), and adult and pediatric patients with elevated white blood cell count but none of the previously mentioned chromosomal translocation]. RESULTS: At least 8 genes show consistently a differential expression profile. The expression of the tumor protein p53 binding protein (TP53BPL), the tumor protein p53 inducible nuclear protein 1 (TP53INP1), the BTB and CNC homology 1, basic leucine zipper transcription factor 1 (BACH1), the protein tyrosine kinase 2 (PTK2), the receptor-like tyrosine kinase (RYK), the proto-oncogene rel (REL), the cyclin D2 (CCND2), and the H2B histone family member S (H2B/S) was between 2 to 7 fold higher in the low-risk pediatric subgroups compared to high-risk pediatric and adult ALL subgroups. The significance of these differences was supported by T-test p-values ranging from 4.6 x 10^-3 to 10^-7. The analysis of the gene expression by the computer algorithms in CLUSFAVOR, resulted in the clustering of these genes in relation to ALL risk groupings. CONCLUSIONS: Our results show that the significance of these specific genes for predicting risk behavior crosses multiple karyotype subgroups. We are currently investigating the interaction between some of these genes that might be part of the same regulation pathway. This suggests that the difference in treatment response between the lower and higher risk forms of the disease might be under the control of a few regulatory networks that could become therapeutic targets to improve ALL clinical outcomes.

Paper presented at the International Symposium on Predictive Oncology and Intervention Strategies; Nice, France; February 7 - 10, 2004; in poster session 892 (Susceptibility genes).