Targeted but risk-adapted therapy to the central nervous system (CNS) has become a prerequisite for successful treatment of childhood acute lymphoblastic leukemia (ALL). Overt CNS involvement is a rare observation in childhood ALL and commonly associated with a poorer outcome when compared to CNS-negative patients. Although there are some known risk factors for CNS involvement of ALL such as T-cell-precursor immunophenotype and high initial white blood cell (WBC) count, the underlying biology leading to CNS disease is poorly understood. We hypothesized that leukemic cells crossing the blood-brain barrier early have distinct biological properties which are reflected in a specific gene expression pattern. To test this hypothesis, we generated leukemic gene expression profiles of 17 patients with CNS involvement using 26 CNS-negative patients as controls. Microarrays containing more than 39,000 distinct cDNA clones (SFGF, Stanford CA) were used. Data were analyzed applying Significance Analysis of Microarrays (

PNAS 2001, 98, 5116–5121
). Our analytical approach identified several differentially expressed genes. To control for possible T-cell contamination of the samples, we purified leukemic cells from four samples and repeated the analysis. Interleukin 15 (IL-15) was one of the genes for which differential expression could be confirmed with an up to 10-fold higher expression in CNS-positive patients. IL-15 is a pro-inflammatory cytokine known to stimulate proliferation and activation of T-lymphocytes and polymorphonuclear cells. Of importance with regard to the CNS, IL-15 was described to be highly expressed in peripheral blood (PBL) mononuclear cells in multiple sclerosis. The high expression of IL-15 in patients with CNS involvement was confirmed in an independent set of 13 CNS-positive and 26 CNS-negative patients by quantitative RT-PCR in a similar range as described above (P Mann-Whitney U-Test = <0.001). Therefore, we postulated that inflammatory processes might be involved in the pathogenesis of CNS disease by increasing the permeability of the blood-brain-barrier and reviewed differential PBL smears of 134 CNS-positive and 335 CNS-negative ALL patients. We observed a significantly higher ratio of immature to total myeloid cells in CNS-positive samples (P = <0.001). Next, we compared IL-15 expression at initial diagnosis of CNS-negative patients subsequently relapsing with CNS involvement (n = 22) to those without CNS disease and being in long-term remission (n = 18). Again, a significantly higher IL-15 expression was detected in patients with CNS relapse (P = 0.01). We conclude that measurement of IL-15 expression could serve as an additional tool to further tailor CNS-directed therapy in children newly diagnosed with ALL. Our results support a role for inflammatory processes, but additional studies are needed to finally elucidate the role of IL-15 and inflammation in the pathogenesis of CNS disease in childhood ALL.

Disclosure: No relevant conflicts of interest to declare.

Author notes

*

Corresponding author

Sign in via your Institution